Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations

1946 Some derivatives of [gamma]-valerolactone Robert V. Christian Jr. Iowa State College

Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Biochemistry Commons, and the Plant Sciences Commons

Recommended Citation Christian, Robert V. Jr., "Some derivatives of [gamma]-valerolactone" (1946). Retrospective Theses and Dissertations. 13388. https://lib.dr.iastate.edu/rtd/13388

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. NOTE TO USERS

This reproduction is the best copy available. UMI

SOME DSinVATIVES OF T'-VALaHOLACfONE

by

Robert V. Christian, Jr,

A Thesis Subraitted to the Graduate Faculty for the Degree of

DOCTOR OB^ PHILOSOPHY

Major Subject: Plant Gheroistry

Approved

Signature was redacted for privacy. tn „ •

Signature was redacted for privacy. Head ooiajor^e^arteeh't

Signature was redacted for privacy. Dean of Graduate CollegIf

loi-ya State College 1946 UMI Number: DP12639

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

UMI

UMI Microform DP12639

Copyright 2005 by ProQuest Information and Learning Company.

All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 QT)365,A^

fABLK OF COirfiCNTS

IHTRODUCTIOri 1 MATKRIAJLS, SPSGIAL APPARATUS AND GSMSRAL PRO- C}aJURES 5

Preparation of T^-Valerolactone by Brovm's Procedure 6 T^-VALiiliOLACTOiJE IM THF. FRI&DEL AND CMFTS HE- AOTICH 7

Review of the Literature 7 jSxperlaental 9

Prepare t3 on of t'-phenyl valeric acid by the procedure of Eljkman 9 Preparet-ion of ethyl 7'~phenylvalerate and diethyl y-jy-' henylene-dlvalerate ... 9 Saponification of ethyl 7<"-phenyl valerate . . 15 Sa'ionlflco.tion of diethyl -vvT-'-T^henylene- dlvnlerate ' 17 PREPAiiATlOH OF 1,4-P£NmUSI)I0L 18

Review of the Literature ...... i8

Reduction of /'-Valerolactone 19

Effect of tenperature upov. the reaction . . 19 Isolation and Identlflcpt.1 on of 3- metbyltetrahydrofuran 20 Reduction of Ethyl Levullnate ...... 21 Reduction of Levullnlc Add 22 PREPARATION OF 1,4-DI-(2-CXAN0STH0XY)-PSt'imWS AND STRUCTURALLY RELATiSD CaiPOUNDS . 23 ill

Page

Addition of Acrylonltrlle to Alcohols and G-iycols 23

Preparation of /3-Allcoxyprorj.ionltril0S .... 25 Procedure 25 Mono functional ^-alkoxjrproplonltrlles of the type ROGHgCHgCN 26 BlfunctIonal 0-alkoxyprop1on1triles of the ty-.e ^{CHgCHgClOg ^>9 FR£t-\41U1'I0N OP 1,4-DI- (3-MlINOPROPOXy) -PENTAM AiJD PIAMINi; 3ALTS 35

Review of Pertinent Literature ...... 35 Reduction of 1,4-Di-(2-oyanoethoyy)- pentane ...... 36 Salts of 1,4-Di~(3-aro3nopropoxy)-pentane ... 38

Procedure 38 Preparption of the salts 39 HYDROLYSIS OF 1,4-DI-(2-CYAK0I{:TH0XY)-PEHTANE AND RELA'fiSD GOMPOaWDS 48

Review of Methods hy Wilch ^-Alkoxyrro- plonlc Acids Hrve Been Prepared ..... 48

Basic Hydrolysis of /3-Alkoyypropionltrllea . . 52 Bftslc hydrolysis of l,4-dl-(2- cyanoethoxy)-pentane 52

Basic hydi'olyslo of di-(2-cysno- ethyl )-ether% ..... 53 Basic hydrolysis of ^-ethoxypro- plonltrile ...... 54 Acid Hydrolysie of p-Alkoxyproplonitriles . , 55 Procedure 55 Iv

Page

Preparation of oonobaslc ^-alkoxy- propIonic aoldfl and their derivatives S7 Preparation of dibasic /S-alkoxy- proplonlc acids and their derivatives ...... 67

Att^pts to prepare l,4-di-(2- carboxyethoxy)-pentane ..... 77 DISCUSSION ...... 84 SUKHABX 92

ACKNOILSD&E^SMTS 94 LITERATURE CITED 95 1

INTRODUCTION

Early In the war, #ien the supply of natural rubber to the United States wae cut off, the proble® of finding synthetic substitutes beosBe esctranely liaportaJit. A possibility con­ sidered at that time WES the use of levulinlc acid (I) as the starting point in the preparation of 1,3-pentadiene (IV) the following series of reactions;

+Hp CHsCOGHgOHgCOOH (^aney Hi)' 0 CO

I II

'.Ho CH3CHOHPCH2P (CuCrO) OH OH (Al^Oj^)

III

CH3CH;CHCH:CHg + CHg:CHCH^CH:CHg

IV

Preliminary studies by Mr. Horace D. Bro^ showed that the process was possible, altiiough not inaaediately practical, and he succeeded in obtaining saaples of rubbarlike material fTOm the copolyroerizatlon of 1,3-pentadlene and aorylonltril©.

The most significant contribution of Mr. Brown*s study was the 2

suoGeesful liquid phase hydrogenatlon of levxillnlc add to

'^-Talerolactone (II) in excellent yield. A elmllar process is the subject of a recently Issued patent (52) and is re­ ported to be under pilot plant investigation by the Monsanto

Chemical Company.

The present investigation was undertaken for the puipose of examining sonse of the reactions of T'-valerolaotone (II)

and 1,4-pentanedlol (III) with view to preparing nei? derivatives

which might prove worthy of further study leading to industrial appliestions. The significance of such e.n investigation be­ comes apparent when the source of levulinic acid is considered.

Levulinio acid results when any hexose sugar, or carbohydrate hydrolyzable to a hexose sugar, is heated in aqueous solution with mineral acid (34, 80, 86). At present, the acid is

manufeotured principally froro starch. The dlscovei^ of new uses for materials derivable from levulinio acid would pro­ vide another approach to the utilization of agricultural by­ products. fhe successful conclusion of such a research pro­ gram, which, of course, might require many years, could re­

sult in a more intelligent exploitation of our natural re­ sources and would be of great importance in the agricultural phases of our national economy. The value of any contribution

to a long-teKD project of this type seems beyond question. The plastics industry probably offers the most important potential outlet for large quantities of moderately priced 3 chemicals. Attention was, therefore, directed to the pos­ sibility of using^-valerolaetone and 1,4-pentanedlol as starting points in the synthesis of dibasic acids and amines from which linear polyanldes of the "nylon" type could be prepared by the usual condensrtion methods (16), Although 1,4-pentanediol has been obtained previously

In satisfactory yield by catalytic reduction (31,40), the ppe- paration of this compotind 'sas re-exaralned Int the hope of improving the yields and adding refineisents to the existing preparative methods.

Prora the many possible ways in which -^^-valerolaotone or 1^4-pentan@dlol could be utilized in the synthesis of bi- functlonal acids or aisines, two were selected for Investiga­ tion. The first line of attack proposed to m^e use of the little knoMi Priedel and Crafts reaction with /'-valerolaotone by roeens of which ElJJcaan prepared "T^-phenylvaleric acid (25). It was hoped that it would be possible to cause benzene to condense with two raolecular equivalents of the lactone to yield a y; •^-phsnylene-dlvalerlo acid or a derivative of such an acid. For the second approach, It appeared that application of the recently reported "cyanoethylation" reaction (6,7,8,14,17,

41,45,51,82,85) to 1,4-pentanediol might yield an interfuediate dinitrile which could be converted to a dibasic acid by hydrolysis, or to a diamine by hydrogenatlon. fhe hydrolysis of the nItrlle prox'^ed to be an entirely new approach to the

Bynthesln of y3-alkoxyprof)lonlc acids and their derivatives. An extension of the study to a series of structurally related cosipounds wrs, therefore, projected ard carried out. 5

MATERIALS, SPSGIiiL APPi-JlATUS MD OSHERAL r^ROGSDURilS

Levullnlc acid used in this etudy was an Eastman Teoh- nlcal product which was distilled ^ vaouo before use,

T'-Valerolaotone was prepared from levullnic acid by th® Hsathod of Broim (4), fVils preparation will be described at the conclusion of this section, Ruslcka'a procedure (71) was followed in the preparation of ethyl levullnate. Alcohols and glycols used in the preparation of ys-alkoxy- proplonltrlles were comroerclally available chemicals. Sest- Bsn Wliite Label chemicals and others of comparfibl© quality were used without further purification* Practical grades were distilled before use and frsotlons boiling over a ranf^e of three degrees or lees ^ere employed In the preparations. The acrylonltrlle was an Eastoan Practical product. In pi-^llmlnary experiments the material was distilled prior to uae but it was fotmd later that the yields of addition pro­ duct did not suffer when it was used \'?ithout purification.

With one exception, the aclde used In the preparation of

the amine salts were conimercially available materials. The reagent quality acids were used directly whereas those of technical or practical grade were purified by suitable methods. The glutarlc acid was prepared from fosmaldehyde and dletliyl jBElonate by a standard procedure (60). 6

Copper-chroralum oxide end Raney nickel catalysts were prepared In the usual way (l, pp. 13, 19). Hydrogenations were carried out In a 500 ml. rockinf^ bomb of standard design

(Parr Instrument Co.). A motor-driven booster pump, manu­ factured by the American Instrument Company, was used to ob­ tain hydrogen at pressures higher than that available from commercial steel cylinders. Stsjidard practices were followed in the reductions. The technique of super-pressure hydro­ genation has been adequately treated elsewhere (1, pp. 29-46).

Preparation of y-Valerolactone by Brown's Procedure

A copper-lined hydrogenation bomb of 3.85 liters' capa­

city was used. The charge consisted of 580 g. (5 moles) of distilled levullnlc acid and 15 g. of Raney nickel. Hydrogen ?ms introduced to a pressure of 700 p.s.l. The temperature during the reduction was 175-200° and the reaction was com­ plete after three hours. Separation of the catalyst distillation of the reaction mixture gave 471 g. (94^) of /'-valerolactone of boiling point 87-90° (8-10 mm.). 7

7-VALEROLACTONE IN THE FRISDSL MD CRAFTS REACTION

Review of the Literature

The use of esters as alkylating agents In the Prledel and Crafts reaction is well known and has been the subject of several published studies (3, 59, 79). The reaction differs from the usual Friedel and Grafts alkylation with alkyl halldes in that slightly more than one mole of catalyst must be used per mole of ester. The reason for this is obvious ^^hen one considers the mechanlsra of the reaction. The first step is the cleavage of the eater by the alianlniOB chloride to yield an alkyl halide and a salt (59), thus?

ROOOR AICI3 ^ RCOOAlClg + RCl The first molecular equivalent ©f catalyst used serves only to bring about this cleavar^e. After fonaatlon of the alkyl halide has occurred, a sllp^ht excess of catadyst is suf­ ficient to cause alfcylatlon of the aromatic component present (79). When the ratio of catalyst to eater Is doubled and the reaction time end t^aperature Increased, the salt can react further (79):

RG00A1G1_ + AlCl, RCOCl + AlOGl d, o I AlCl, o 8

The active complex thus fonaed is capable of taking part in an acylatlon reaction smd ketones result. However, a proper choice of conditions raskes it possible to confine the reactim to alkylation (59), Lactones, as cyclic esters, should be expected to undergo a similar reaction. This is indeed the case, although Friedel and Crafts reactions with unhalogenated, aliphatic lactones have been reported by only one investigator. Durinf- the course of studies on seven 7^~lactones (25-28), Eijkman reported (25) the preparation of ^^-phenylvaleric acid and /"-tolylvaleric acid. These acids resulted from the reaction, in the presence of aluminum chloride, of 7«-valerolactone with benzene and toluene, respectively.

Eijkman did not report the yields in his preparations, nor did he determine the position of substitution in the re­ action with toluene. In neither case was proof presented to indicate that the T^substituted valeric acid was the result. This point assumes significaiice in view of the fact that isomerlzation freouently accoropanies Priedel and Crafts alkyla- tlons (7S, pp. 94-97). 9

Experimental

Preparation of y-phenylvalerlo acid by the procedure of El.lkinan (25)

To a well-stirred suspension of 80 g. (0.375 mole) of alumlniaa chloride In 60 g. {0,77 mole) of benzene, there was added dropwlse, 30 g. (0.3 roole) of ^-valerolactone. A vigorous reaction wllii evolution of hydrogen chloride began at once. The mixture was cooled occaalonally to keep the re­ action imder control. After all the lactone had been added, the mixture was heated in a water bath for 25 minutes. It was cooled and poured Into a mixture of 100 g. of ice and 100 g. of concentrated hydrochloric acid. The benzene layer was diluted with ether, extracted with dilute hydrochloric acid, ?^ashed with water and distilled vacuo# There was obtained

32.5 g. (61^) of colorless, oily liquid of boiling point 123-126° |0.5 mm.). It was Insoluble in water, but dissolved in sodlura bicarbonate solution wl13i evolution of Ga,rbon dioxide, Eljtoian {05) used the same quantities of reactents in his preparation, but he added the cats.lyst to a solution of the lactone in benzene,

Preperation of ethyl •Y--phenylvalerate and diethyl Y-Phenylene- dlvalerate'

In experiments in which attempts were aade to effect 10 dlalkylation, the crude reaction mixture ?/as esterified to facilitate purlficetion by distillation. The esterification procedure is as follows; The crude acid ?ms mixed with 2.5 to 3 tljnes its volume of absolute ethanol and about one-tenth its Yolume of concentrated sulfuric acid. After being re- fliixed for three hours, the mixture was poured into water end the esters were extracted with ether. The ether solution was extrrcted with saturated sodium biCc?,.rbone.te oolution, washed with water, dried over sodium sulfate and distilled. Yields

7/ere then determined upon the basis of the qusntities of ethyl esters obt&lned.

(I) A mixture of 52 g, (0,18 mole) of 7^-phenylvaleric acid, 20 g, (0.2 mole) of -^-valerolactone and 100 ml, of carbon disulfide was cooled in ice and stirred vlgozHDuely while 53.2 g, (0.4 sole) of aliaainum chloride was added in small portions durin^j 20 minutes. The ice bath was removed end stirring was continued at room temperature for 24 hours. The mixture was hydrolysed with ice and hydrochloric aold. After reeoval of the carbon disulfide layer, the aqueous solution was extracted with ether. The combined ether and

carbon disulfide extracts were washed with dilute hydrochloric acid and then with water. Evaporation gave 31 g. of oil which was esterified with ethanol.

The mixture of esters was fractionally distilled to

yield 14 g. (38^) of ethyl 7^-phenylvalerat© of boiling point 11

100-103° (1 mm,) and 7 g. {1Z%) of diethyl x-phenylene- dlvalerate of bollinf; point 183-190® (1 nra.) (II) A suspension of 93.1 g. (0,7 mole) of alUBinum chloride in 150 ml, of carbon disulfide was vigorously stirred while a solution of 23.4 g. (0,3 mole) of benzene and 60 g, (0,6 sole) of T'-valerolactone an 50 ral, of carbon disulfide

•was adcted dropwise. The mixture becpine warm and hydrogen chloride was evolved. After beinp; stirred for five lioura at room temperature, the reaction mixture -was heated to gentle reflux for 30 minutes, and then allowed to stand overnight.

The gummy complex was broken up and the mixture was stirred and refluxed for 40 minutes. Hydrolysis, effected in the usual manner, was followed by ether extraotion. Part of the product was lost when the ether solution was spilled. Evaporation of the ether extract gave 21 g* of an acidic gum which was esterified directly.

Upon distillation, the ester mixture gave 4,5 g, (8^)

of ethyl 7^-phenylvalerate boiling at 104-110® (l ma,) and 10,5 g, (11^) of diethyl 7^ 7^-phenylene-divalerate of boiling point 185-195*^ (1 mm,),

Biere leraained 5-6 g, of clear brown residue. This slowly dissolved when refluxed with 50 ml, of 25^ sodi\M hydroxide

solution. Acidification of the clear solution caused the precipitation of a gum which ^as extracted with ether. Evaporation of the ether extract gave 5 g, of an acidic glass 12

Of neutral equlvelent 154. Attaapts to purify it by reci^stal- lizatlon were fruitless. It is considered likely that this material waa the result of trialkylatlon. If this be the case, the yield of trialkylatlon product (based upon /•-valerolactone) was 6^. (III) A ffiixtui^ of 40 g. (0,4 sole) of T'-valerolaotone and 31,2 g. (0.4 raole) of benzene was cooled to 0° fmd stirred while 60,3 g. (0.5 mole) of altaalnuia c^iloride was added in small portions over a pBrlod of 30 minutes. The halld© die- solved as it was added and the solution became llr^ht brown in color and viscous. No hydrogen chloride '^?as evolved until all tho catalyst had been introduced. Then an explosively violent evolution of hydrogen chloride began and the reaction went rapidly out of control. The product was lost» (IV) A solution of 40 c- (0.4 aole) of 7^-valerolactone and 31.2 g. (0,4 mole) of benzene In 75 ml. of nitrobenzene

was cooled in a freezing mixture and stirred wh-le 66,5 g, (0,5 mole) of aliaiiniaB chloride was added in small portions

during 20 minutes. The catalyst dissolved and the mixture took on a yellow color. Ins-Smuch as no action was observed, the freezing mixture wps removed. After about 10 minutes a slow evolution of hydrogen chloride began.. The mixture was

stirred at room temperature for 18 hours and then heated at o 80-90 for 2 hours, !i:^e usual hydrolysis waa followed by steam distillation to remove nitrobenzene. The residue was 13 teken up In saturated sodium bicarbonate solution and extracted witii ether. Acidification sund ether extraction gave 27 g, of thick, brown oil. This was converted to a mixture of ethyl esters. Distillstion gave 13 g. (16^) of ethyl 7^-phenylvalerate boilinf^: at 104-108® (l mm.) and 11.5 g. (17,^) of diethyl 7^ ^^-phenylene-divalerate of boiling point 183-187° (l mm.). (v) A solution of 40 g, (0.4 mole) of /--valerolaotone end 31.2 g. (0,4 mole) of benzene in 70 ml. of carbon di­ sulfide was cooled in ice and stirred t^hile 66,5 g, (0,5 mole) of aluminum chloride was added in six portions over a period of 20 minutes. Very little hydrogen chloride was evolved until all the catalyst had been Introduced. Stir­ ring was continued for two hours at 0° and then for four hours at room temperature. After standing overnight, the o mixture was stirred and heated at 90-100 for 40 minutes. Hydrolysis and ether.extraction gave 60 g. of brown syimp which was esterified*

Distillation of the mixture of esters gave 31 g. (38^) of ethyl 7--phenylvalerate of boiling point 97-104° (l mm.) and 32 g, (48^) of diethyl •7'-phenylene-dlvalerate boil­ ing at 175-183° (1 rom.). The Information obtained in the foregoing experiments is summarized in Table I. 14

Table I

Alkylatlons with ^valerolactone

Reactants Aluminum EXD. Lactone Benzene Solvent ciilorlde E No; (moles) (moles) (ml.) (ujoles) Products

b I 0.2 none 100 0.4 CO CD

II 0.6 0.3 150 0.7 8^ A ° CS2 11^ B III 0.4 0.4 none 0.5 d

IV 0.4 0.4 75 0,5 16g A CeHgNOg 17^ B Y 0.4 0.4 70 0.5 58% A GSg 48^ B

The yields were calculated from the amounts of ethyl T-phenylvalerate (A) and diethyl 7^,7--phenylene-dlvalerate (B) recovered subsequent to esterlficatlon of the crude reaction products,

^ T'-Phenylvalerlc acid (0,18 mole) used instead of benzene. °Slx per cent of a substance that may have been trlalkyl- atlon product was obtained in this experiment. • dThe reaction went out of control and the products were lost. 15

Carefully purified ethyl T'-phenyl valerate had the follow­ ing physical constsnts: boilinn point 108-109° (l ma,), ?5 » d ^ 0.9850, n 1.4898, Levene and Marker (55) report boil- " ^ ~O D OR ing point 112 (1 lam.) end d ^ 0.985 for the levo-fonn of

the ester. The physical oonstante for pure diethyl divalerate were as follows; boiling point 165-169° (0.5 mra,), PS PS d . 1.0212, n „ 1.4914, — 4 » jj Anal. Calcd. for C2oH3o®4* saponification equivalent, 167; C, 71.9; H, 9,04. Found: Mr^j, 94.84;

saponification equivalent, 169, 171; C, 72.2; H, 9.37,

Saponification of ethyl nhenylvalerete

A solution of 30.9 g. (0.15 mole) of ethyl -^-phenylvalerate and 14 g. (0.35 mole) of sodium hydroxide in 60 ml. of 50^

ethanol was refluxed for three hours. The mixture was worked

up in the usual way to yield 23 g. (86^) of T'-phenyl valeric o ?5 15 acid of boiling point 120-121 (0.5 mm,), d 1.0474, d PS 15 1.0556, n'^jj 1.5113. Eijkiaan (25) reported & 1,0554 for h-is preparation* For 7^-phenylvaleric acid prepared from 25 3-phenylbutyl bromide, Levene and Marker (55) reported d"^^ 1.040, 16

p-Bromophenao.yl -^-phenylvalerate« In accordance with the procedure of Skrlner and Puson (76), 1 ml, of /'-phenylvaleric acid end 1 r?. of p-broroophenacyl bromide gave the desired derivative as large, shlnino:, xvhlte leaflets. The melting, point wfs 76® after recrystallizatlon from dilute methanol.

Anal« Calcd. for G^gH^gO^Br: Br, 21.3. Pound: Br, 21.1. Proof that -^'-phenylvaleric a.cid prepared by the Frledel and Grafts reaction was identical with the compound prepared by other methods (50, 55, 57, 84) was afforded by ring closure to the Imown 4-raethyl-l-tetralone. The previously reported seialcarbazone was then prepared from this compound.

4~Methyl"l--tetralone, In accordance with the procedure of Kloetzel (50), 10.5 g. of 7^-phenylvaleric acid was heated with 54 BJl. of 80^ sulfuric acid for 2.5 hours at 95-100*^. The reeulting solution was poured into 160 ml. of water and this was extracted witti ether. The ether extract was washed Mth nonaal sodium hydroxide and then with water. Distilla­ tion of the dried ether solution gave 5 g. of 4-methyl-l- o 19 tetralone of boiling point 103 (l mm.) and n ^ 1.5610,

CoiDp&rable physical, constants for this cotspound are reported o l9 to be boilinp; point 110-111 (l mm.) (50) and n 1.5620 " D (84).

4-Methyl-l-tetralone seaicarbazone. The seralcarbazone was prepared from 1 g. of 4-Biethyl-l-tetralone using the procedure of Shriner and Fuson (76, p. 145). Recrystallizatlon 17 from dilute ethanol gave tiny, shining, 'shlte leaflets of o melting point 207-209 , The melting point of this compound has been reported as 210° (57), 204° (84) and 209-211 (50),

Saponifloatlon of diethyl Y» -y^^Phenylene-dlvalerate

To a Solution of 16 g. (0.4 mole) of sodliau hydroxide in 60 ml, of 50^ ethsjiol there was added 30,1 g, (0.09 mole) of diethyl y, -^-phenylene-divalerate. The mixture was boiled under reflux for three hours. It was worked in the customary fashion to yield 24 g, 6f a light browi srvm of neutral equivalent 149, The calculated value for the neutral equivalent of 7-,-^-phenylene-divalerlc acid is 139, The gum forsed a sticky, semi-crystalline mush on long standing. Attempts to obtain the compound in a satisfactorily oi^stal- llne state failed. 18

PREPARATION OF 1,4-PENTANSDIOL

Review of the Literature

1,4-Pentanedlol \?as first prepared by the reduction of 3-acetopropanol in aaueous solution with sodium amalgeiB

(18, 32, 56). More recent ?/orkers have usiially employed '^-vslerolactone as the starting material. Chemical reducing agents that have been used are sodium and ethanol (?5),

ecetlc acid and a toluene suspension of sodltaa (54), and aodlum amalgaiD In 10^ sodliaa carbonate solution (44), These methods of preparation have various disadvantages. Working In aqueous solution complicates the recovery of the glycol because of Its high solubility In water. For example, Colman and Perkin (18) reported that 20 to 30 ether extrac­

tion a are insufficient to recover all the glycol end that careful and efficient fractionation Is necessary to separate the product from traces of alcohol and water. Furthermore,

it would appear that strongly basic reducln^^ a^;^ents may have a deleterious effect upon the lactone. Thus, Buckel

end Gelsroth (44) obtained only a ZO^ yield of 1,4-pent8iiedlol

and 45^ of higher boiling material in their preparation, Ihe advent of copper-chromium oxide as a catalyst for the hyfiro£enation of esters to the corresponding alcohols

has made 1,4-pentanedlol more easily available. Folkers and 19

Aditlna (31) have prepared the glycol In yield from y-valerolactone. The reaction VJBB carried out without solvent st 250° and 200 atmospheres* pressure. !Riey also Isolated 9^ of n-aiayl- alcohol. Under the same conditions of temperature and pressure, ethyl levullnate In dloxane or ethanol solution gave 60^ of 1,4-pentenedlol {40)»

The catalytic methods have niaaerous advantages. Glean, easily purified reaction lalxtures are obtained. Working without solvents slopllfles the recovery of the products. The yields are usually .o^ood or can be made so by proper selection and control of the experimental conditions.

Reduction of 7^-Valerolactone

Effect of temperature upon the reaction

(X) The chsrge consisted of 100 g. (1 roole) of-^-valerO- laotone and 6 g. of copper-chromlu® oxide. The Initial hydrogen pressure was 5400 p.s.i. Reduction began at about S25 degrees with a gauge pressure of 4520 p.s.l,, proceeded rapidly at

240-250® and was completed 40 minutes after hydrogenation commenced. The maxlmiaB temperature reached was 254®. Distilla­ tion of the filtered reaction mixture gave 86.5 g. (83^) of

1,4-pentanediol boiling at 125*^ (16 ims.). The crude reaction mixture had an odor of 2-iaethyltetrahydrofuran but none could be Isolated by distillation at afeoospheric pressure. P.O

(II) A mixture of 100 f-;. (l mole) of -T^-valerolactone

and 6 g. of copper cbromltam oxide was hydrogen a ted with an

Initial hydrogen pressure of 3120 p»s.i. AbBoxn>tlon of o hydrogen began at about 220 and stopped 35 minutes later. o The aaxlmm> tensperature reached wc.s g?o , Upon distillation of the filtered product a fr'.rctlon was obtained which boiled o at 73-88 under atmos-'heric pressure. TnlB contained ^ater

had a strong odor of 2-aethyltetrahydrofuran. Ihe residue was dlatllled under diminished pressure to yield 10.5 g.

'-•'Oilinr frow 40^ to 1?A^ (16 nnn.) snd 78 fj. (75;l) of 1,4-

pentanedlol of bo11 In • - oint Ig^-lS?*^ (16 Rim.). (III) -^-Valerolactone (100 , 1 mole) reduced in the presence of 9 g. of copper-chromlutn oxide under an

initial hydrogen pi-'essure of 5300 p. 8,1. The maxiroiro tero- o perature attained 2^0 . DiBtlllatlon of the filtered islxture gave 32 g. of mobile liquid of boilln. ' ranrje, 70- o 120 . This fraction contained a little wnter and a stron/^ odor of 2-methyltetrahydrofuran waQ noted. The remaining rocterial was distilled i^ vacuo. A fraction boiling from o o 48 to 124 was discarded. 1,4-Pentanedlol was obtained as 33.5 g. (52;') of colorlees, viscous liquid of boiling point, 124,-127° (16 mm.).

Isolation and Id en 11f 1 c e.t i on of 2-ine th}/"!te t rahyd rof ur an

The low boiling fraction® (b.p. 70-100° at atmoapherlc

pre-ssui'e) froti? several preparations of 1,4-pentrinediol were collected and dried with anhydrous -iOtassium cerbonr-te for 21

several flays. The material was refluxed \?ith sodium for three hours and finally distilled from sodium. The material which boiled at 78-80° had the characteristic odor of 2- sethyltetrahydrofuran. Other physical constants were found on 20 to be d^ 0.8548 and n 1.406. Zellnskll and Shulkln (93) ~ 4 - D give the following properties for 2-ffiethyltetrahydrofuran: boiling oolnt 78.5-80® (746 mm.), d"^ 0.8552, and 1.407. — 4 — jj A solid white cake remained in the distilling flaiSk. Tills wes cautiously treated with araall portions of water until it dissolved. A considerable quantity of oily liquid separated fron the aqueous solution. It is quite possible that this was a mlicture of amyl alcohols which had been converted to the Glkoxides by treatoent with sodium. It will be recalled that Folkers and Adklns (31) obtained an 8^ yield of n-amyl elcohol Ir addition to l,4~pentanedlol from the copper- chroiDium oxide catalyzed reduction of -^-valerolactone.

Reduction of Ethyl Levullnate

A rolzture of 100.8 g. (0.7 mole) of ethyl levullnate ajid 8 g. copper-chromliaa oxide IVP.B hydrogenated at an Initial pressure of 3100 p.s.l. The reduction anparently took place li:: two stages. An extremely rapid absolution of hydrogen took place at about 160^. Rapid reaction was again observed at 230-240^. Absorption of hydrogen stopped 25 minutes after 22

the reaction "began. The reaction mixture was filtered end distilled l|j vaouo to yield 52 g. (72,^0 of 1,4-penteJiediol

boilinsj St 123-125°AS mm.

Reduction of Levulinic Acid

Brown (4) hydrogenated levulinic r.cid over oopper-

chrosiuiB oxide at 100 atrsospheres end 270® to obtain &2$

of ^-vslerolactone and 21^ of 1,4-pentsiiediol. It was felt

that an incresse in pressure might improve the yield of the

(jlyool at the expense of the lactone.

The charge consisted of 92.8 g, (0,8 mole) of distilled

levulinic acid and 8 g. of copper-chromiUEi oxide. 51ie initial hydrogen pressure was 200 atmospheres. Reduction, which begaj'? at 190° with a'^aur^e pressure of 267 atmospheres, toofc place in two steps. About 0,9 sole of }grdrogen was o o telcen up at 245 and an equal quantity at 300 . The total

tiroe for coopletlon of the reaction (ss evidenced by cessation of hydrogen absorption) was one and one-t)iird

hours and the maxioum temperature reached was 304*^. Fractionation of "ttie filtered product ga.ve 9 g, (11^) of-y^valerolactone (b.p. 89-92° et IS m n,) and 36.5 g. (44,^) of 1,4-pentanediol (b.p. 122-124° at 15 ram.), A forerun of

18 g, (b.p. 80-103° at atmospheric pressure) contained water and had a strong odor of 2-methyltetrahydi^furan. 23

PREPARATION OF l,4-DI-(2-CyAN0£TH0Xy)-PENTANS AND STRUCTURALLY RELATED COMPOUNDS

Addition of Acrylonltrile to Alcohols and Glycols

The first observation of the base-catalyzed addition of acrylonltrile to the hydroxyl group of alcohols to form a 2-cyanoethyl ether was probably that of Koelsch (51). Credit for prior publication must, however, go to Bruson and Riener (14). Koelsch (51), In the course of studies on the Michael condensation of acrylonltrile with various active methylene compounds, observed that when the reaction was carried out in ethanol solution, ^-ethoxypropionitrlle was isolated in hif^h yield. Bruson and Riener (14) voidertook Bystematlo stu(31es of the reaction and extended it to a nuiaber of glycols.

Isolating, of course, the expected bis-(2-cyanoethyl)-ethers.

It has been shown further, that the reaction yields the 2- cyanoethyl ethers of ether alcohols such as 2-iBethoxyethanol (6, 8, 82), alicyclic alcohols such as cyclohexanol and re­ lated compounds (7, 82), and arallphatic and heterocyclic alcohols such as benzyl alcohol (82) and tetrahydrofurfuryl alcohol (6, 8, 82).

The reaction has been carried out with a ntmber of siaple aliphatic primary and secondary alcohols (17, 45, 82), in each 24

case the desired 8-cyanoethyl ether being obtained. Acrylonl- trile has been shown to react with water tinder the conditions of the reaction to yield di-(2-cyanoethyl)-ether (14, 41), prestunably by way of ethylene cyanohydrin as the intermediate compound (14). A recent patent (85) reports the reaction of acrylonitrlle with aqueous formaldehyde in the presence of sodium hydroxide to yield 2-(hydroxsTOethoxy)-propionitrile and di-(2-cyanoethoxy)-iBethane. The formaldehyde apparently reacts as the hydrate.

The addition of acrylonitrlle appears to be fairly general with primary and secondary alcohols and high yields of the ^-alkoxypropionitriles are obtFined. Tertiary butyl alcohol, however, has been showi (5, 12, 13, 14) to be inert toward aciT-lonitrlle in the presence of bases and ceii be used as a solvent In the reaction. It has been observed (82) further, that secondary alcohols give lower yields than primary alcohols.

No reaction talces place between alcol^ls and acrylonitrlle in the absence of a catalyst (82). oodium alkoxldes (14, 51), 40^ aqueous potassium hydroxide (14), 38-40^ trimethylbenzyl- amiDoniiaa hydroxide (14, 82), and metallic sodium (17) have been employed to catalyze the reaction. The reaction undoubtedly reaches an equilibrium condition,

ROH T CHG; CHGN ROCHGCHGCJN, inasmuch as unreacted starting materials can be isolated after 25

neutralization of the catalyst (82). The fact that Iso- propyl alcohol gives a louver yield of addition product than n-propyl alcohol has sugrceated that the equllibritffla position is less favorable In the ccse of a secondaiy alcohol (82).

Preparation of y3-Alkoxyprop ion it riles

Early attempts to convert 1,4-dl-(2-cyanoethoxy)-pentEn0

to the corresponding acid indicated the wisdcxa of a systematic investigation of the hydrolysis of y3-alkoxypropionitrilee. Accordingly, in addition to the preparation of l,4-di-(2~

cyanoethoxy)-pentire, the synthesis of a ni;BBber of structurally similar compounds is d.eBcribed.

Procedure

With the exception of ^-ethoxypropinnitrlle, the mono- functional ^-alkoxypropionitriles ^ere prepared In accordance with the procedure of Uterinohlen (82), The alcohol (1 to 1.5 moles) wes stirred under reflux with 3 ml. of aqueous 40^

potassitsB hydroxide solution while a molecularly equivalent quantity of acrylonitrile was added at such a rate that the temperature did not rise above 35-45°. The reaction vessel was cooled in water when necessary. The mixture was stirred for several hours after all the acrylonitrile had been added, acidified with glacial acetic acid and distilled iUi vacuo. 26

The general method of Brueon and Rlener (14) was used for the preparation of the cyanoethyl derivatives of the glycols.

This was essentially the asme as that given above except that the eiaount of catalyst used was 5-7/^ of the wei^t of the glycol and two molecular equivalents of aorylonitrile were added. The mixtures were neutralized with hydrochloric acid. Any variations in the general procedure are given in connection

?;ith the individual experiments.

Monofunotlonal ^->alkoxypropiQnitriles of the type ROCHpCHoCN

yS-Ethoxyproplonitrile* In accordance with the procedure of Koelsch (51), a piece of sodiiM the size of a pea was dis­ solved In 101 g. (2,2 moles) of absolute ©thsnol. To the resultinfi solution, 106 g. (2 moles) of aorylonitrile was

added in six portions with frequent shakin^^. The addition" required .thirty minutes and the mixture was kept cool by

imaerslon in ice-water. The flask was stoppered pnd allowed to stand at room temperature for a day. Neutralization with one ml. of glacisl acetic acid follo%ved by distillation gave

187 g, (94*5^) of ^-ethoxypropionitrile of boiling point

168-171®. Koelsch (51) reported a yield of &9% and a boiling point of 170-173°*

y3-n-Propoxypropionitrile. A mixture of 90 g* (1.5 mole)

of n-propyl alcohol and 3 ml. of 40^ potassiiM hydroxide was 27

tre^ited with 79,5 g. (1,5 moles) of acrylonitrlle. Heutraliza- tlon of the catalyst by means of glaoial acetic acid followed by distillation g&ve 142 g* (84;?;) of ^-n-propoxypropionltrile o o , 20 of hoilin.r: point 37-89 (24 ) or 84 (19 ran.), d 0,8988, 20 ^ PO , ~ ^ 0.9006, 1.4131.

Anal* Galcd, for CqHtj_OM; N, 12.4. Poimd: N, 12.2. ;g-leo-Propoxyoropionitrile. Acrylonitrlle (79.5 g., 1.5 moles), 90 g. (1.5 moles) of Iso-propyl alcohol and 3 ml* of Of telyat r-ave, subsequent to aoidificatlon and distilla­ tion, 125 g, (74?j) of ^"1 so-'-^roDOxypropion 1 tr 11 e boilinr- at 81-82*^ (24 ram.). Uterraohlen (82) obtained a 69j' yield of o , . product of boiling point 82*5 (25 mra. ).

i3-n-Butoxyr)ropioni trile. In like maun or, 74 g. (l mole) of n-butyl alcohol and 53 g. (1 mole) of ecrylonitrile cave 108 g* (85:;S) of ^-n-butoxyproplonitrlle boiling at 98^1 (20 rm,)t Uterroolilen (82) obt-r^ined an 86;^ yield and found the o , boiling point to be 98 ( 20 mro.)* iso-Butoxypropionltrile. Prora the reaction of 74 g. (1 mole) of l8o~butyl alcohol with 53 g. (1 mole) of acryloni- trile in the presence of 3 ml, of 40^ rjotassiua hydroxide, there was obtained, subse^naent to acidification and distilla­ tion, 103 g. (81J®) of p-1 SQ-butoxyproplonitrile of boiling nolnt 08-.'0° (19 ram.) or 91° (20 ram,), d^^ 0.8821, d^^

0.3836, n"^ 1.4143. Ansa. Galcd. for Gr^^gOM; N, 11.0. Pound: N, 11.1. 28

^-sec-ButoxyproPlonltrlle, When 53 g. (l mole) of acrylonitrlle was added to 74 g. (1 mole) of sej-butyl alcohol in the presence of 3 ml. of catalyst, a sluggish re­ action took place. Very little heat was evolved no that cool- Inr; was not neoeseary to keep the temperature below 40°. The reaction ral:Ktur0 was worked up In the usual way to yield 100.5 g. (79jC) of yS-jiec-butoxypropionitrlle of boiling point 90° Of) 20 Pfs (19 mm.), d"^ 0.8880, d 0.8896, n 1.4156. 4 -go " D

Annl. C&lcd. for Gj;H^20Mj N, ll.O. Foundl N, 11«3«

^-Aso-Amyloxypropionitrile. In the customary manner,

85 g. (0.965 raole) of iSQ~amyl alcohol and 53 g. (1 mole) of acrylonitrlle r;ave 112 g. (82^) of &-iso-aayloxypropIon 1trile of boiling point 99° (13 ms.), d^® 0.8818, d^° 0.8834, 20 n^^ 1.4218.

Anal. Calcd. for GgH^gON: N, 9.93. Found: N, 10.1. g'-seo-AroylOiKyproplonltrile. Acrylonitrlle (63 {%, 1 sole) was added dropwise to a vigorously stirred mixture of 88 g.

(l mole) of seo-'amyl alcohol aiisd 3 ral, of potassium hydroxide. The reaction was slow in starting so 3 ml. mors of the 40)1 potassium hydroxide solution was added after half of the acrylohitrlle had been Introduced. External cooling was not necessary and the temperature rose to only 42°. lOie mixture i?as stirred for seven hours at room temperature and then allowed to stand overnight. Tlie potassim hydroxide was neutralised by the addif on of 6 ml. of glacial acetic acid. 29

Distillation under diminished pressure -9 S» (?0^) of o 20 colorlesB liquid, boiling; point 98 (16 nsB.), d ^ 0.8844, OA OA 0.8862, 1.4205, —20 * D Anal. Calcd, for CqH3_50N: N, 9.93. Poimd: N, 9.85.

^-Allyloxypropionitrile. The usual procedure was followed in the reaction of 58 g. (l mole) of ellyl alcohol with 53 c. (l mole) of aorylonitrlle. "Phe yield was 94 g. (85,€) of product of boilinf^- point 95® (24 mm.) or 92 (19 inm.), QQ OA 20 d'^^ 0.9380, d 0.9396, n „ 1.4330. - 4 D Anal. Calcd. for CgHgON; N, 12.8. Found: N, 12.4.

^(2--Methoxyethoxy)-proplonitrlle. In accordance ?/lth Uterniohlen'8 procedure (82), 96 g. (1.25 soles) of ethylene

glycol monomethyl ether vras allo-jed to react/^jith 66.2 g.

(1.25 raoles- of aorylonitrlle to yield 136 g. (85^) of the desired compound boiling at 107-109° (13 mffi,), Uteraohlen (82) reported a yield of &7% and the boiling point as 98~ 100° ( 9 mm.) in his preparation.

®iey3-alkoxyproplonitriles are colorless liquids with faint, pleaeant odora. Kie phyaical constants of some of th^

are corapared In Table II, together v/lth explanatory notes.

glfunQtlo^^ yg-gll^OxyprQPlOffitrlX^a Stl^ toS Y(GHoCHoCNio

Pi-(2~cyanoethyl) - ether. "Ehe reaction of ?1 g. (1 mole) of ethylene cyanohydrln with 56 g. (1.05 rnoles) of acrylonltrile

gave 77 g, (62^) of di-(2~cyanoethyl)-ether boilinc G.t 152- 30

156^/3 ram* fhe boiling point io re jorted (14) to be 161-

162^ (5 mm*).

Table II

Physical constants of some ^-alkoxypropionitrlles RGCHpCHgCIl

Bolllnr^ point 20 20 R d n *^0. roa. - 20 - D

Methyl^'^ 85.5 49 0.9420 1.4032 Sthyl^ 77-'/a 25 0.9062 1,4068

-Propyl 87-89 24 0.9006 1,4131 a Xso-Vvoml^ 82.5 25 0.8969 1.4089 n-Butyl 98 20 0.8865 1,4180

Iso-Biityl 91 20 0,8836 1.4143 sec-Butyl 90 19 0.8896 1.4156 iso~M.vl 99 13 0.8834 1.4218 sec-feyl 98 16 0.8862 1.4205 Allyl 95 24 0^9398 1.4330 s. 2-Me thoxye thyl 98-100 9 —

^Previously prepared. The physical constants are those given by Utermohlen <82)#

^Not prepared In the present study. 31

1,2-Di-(2»eyanoethoxy)~ethane» The reaction of 62 g. (1

mole) of ethylene glycol with 117 g. (2.2 moles) of acrylonitrlle, carried out in the usual way (14),gave 63 g. (37^) of product o of boiling point 158-162 (l mm.). Bruson and Rlener (14) give 158° (2 mm.) as the boiling point.

1.5«-Di'-(2'-cyanoethoxy)-propane. Triraethylene glycol (45,6 g., 0.6 mole) reacted with 63.6 g. (1.2 moles) of acrylonitrlle to give 98 g. (90^) of the expected dinitril®

having the boiling point 165-17^*^ (1 mm.). The previously reported boiling point is 165° (1 mm.) (14).

Pi-( z-fz -cyanoethoxy7-ethyl)-ether. In the usual way,

100 g. (0.94 mole) of diethylwie glycol was allowed to react with 106 g. (2 moles) of acrylonitrlle to yield 129 g. (65^) o , of the desired compound of boiling point 190-200 (1-2 ram.). o The boiling point given in the literature (14) is 190 (1 mm,). 1,4-Dl-(2-oyanoethoxy)-pentane. (I) A mixture of 62 g.

(0,6 mole) of 1,4-pentanedlol and 3 ml. of 40^ aqueous

potaeaium hydroxide was vigorously stirred and cooled in water while 69 g. (1,3 mole) of acrylonitrlle wss added dropwise at o Buch a rate that the temperature did not rise above 35 .

Stirring was continued at room temperature for five hours. After standing overnight, the mixture was diluted with twice Its vol\®e of ether and extracted twice with 60 ml, portions of nonaal hydrochloric acid and once with water. The washings were extracted once with 50 ml. of ether. The combined ether 3g

extracts were dried with anhydrous sodiiaa sulfate and dlstilied mder diainiehed pressure. After a forerun of 7 g,, 104 g.

(83^) of l,4-di-(S~oyanoetho3cy)-pentane was obtained as a o , 28 colorless liquid of boiling point 157 (1 mm.), a 1.4478, ^28 , ^ d 1•005* - 4 Anal, Calcd. for Found? M, 13,5,15,4,

(II) To a aechanically stirred mixture of 61 g. (0,59 aiole) of 1,4-pentanediol and 3 ml. of trioethylbenzylasBionim i^droxide (38^ aqueous solution), 66,2 g. (1,25 moles) of aorylonitrile was added dropwise. Very little heat was erolired. After five hours* stirring the mixture was clear and colorless, had a strong odor of aorylonitidle, and was completely soluble in water. Upon the addition of 3 ol, of 40^ potassiiffli hydroxide the laixture turned light yellow and o the temperatiire rose rapidly to 50-60 ^ It was cooled in water and stirred overnight, Heutralization of the catalyst, followed by distillation, gave 86 g* (70^) of l,4-di-(2- cyanoethoxy)-pentane, (III) fo S3 g, (0*51 mole) of 1^4-pentanediol there was added a solution prepared by dissolving a piece of sodit® the

size of a pea in a little absolute ethanol. It was stirred and cooled in water while 29 g, (0,55 mole) of aorylonitrile was added dropwise. After ten hours* stirring at rooii temperature 26,5 g, (0,5 sole) more of aorylonitrile was added 33 and stirring was continued for three hours, fhe mixture was allowed to stand at room temperature for eight hours. Ether WES added and the solution was extracted with dilute hydro­ chloric acid, then with sodiira bicarbonate solution, mid finally dried over sodium sulfate. Distillation gave 70 g,

(66^) of l,4~di-(E~cyanoethoxy)-pentane. l,4-Di-(2-oyanoethoxy)-pentsne is an odorless liquid, soluble in organic solvents, but not appreciably soluble in water* Attempts to prepare a solid derivative of tdbie dlnitril© by treatffient with alkaline hydrogen peroxide (14, 21, 69), condensation with thioglycollic acid and dry hydrogmi chloride (19), or conversion to a phlorophenone (42), yielded only oils. A solid iainoester hydrochloride was obtained, however, by the usual methods (66, 67), Bis~(ethyliminoester hydTOchloride) of l»4~Di~(2«-cyano-'

eth3roxy)">peatatte> A ffllxtu3?e of 42 g. (0.2 aole) of 1,4-di'-

(2-cyanoethoxy)-pentane and 23 g. (0.5 mole) of absolute ethanol (distilled from sodiiai ethoxide) was treated at

-10^0, with excess, dx^ hydrogen chloride. The resulting thic^ syrup set to a sass of white ciiystals after standing for a week in the refrigerator. After rffluoval of excess ethanol smd hydrogen chloride in vaouo the piH>duot weighed 75 grams (theoretical yield). A sample taken for analysis was pulverized, washed well with anhydrous ether and dried in 34

a yaouua desiccator over sulfuric acid. The result was o a white powder which melted at 103-104 with decoaposition and gas eyolution,

Anal» Galcd. for C, gH^gO^NgClg: 01, 18,89, Fouadj 01, 18.84, 18.94. 35

PREPARATIOM OF 1,4-.DI-(3-AMINOPROPOX3r)-PENTAIIE AHD DIAMIME SALfS

Review of Pertinent Literature

Nitrlles are readily hydrogenated to prlaary amines over Raney nickel catalyst (l, p. 53), However, an Important side reaction is the foriaatlon of seoondary amines at the expense of primary aalnes (43, 74, 82, 87, 88, 89)• The process can be siasmarized by the following equations?

RClli • HG RCHjHH RCHlNH 4. Hg >-

R0HJNH RCHgKHg + Hg

i- (RGKg)gNH + HHg

If the reaction is carried out in the presence of amfflonia, the last reaction is suppressed and excellent yields of primary aiHine can usually be obtained (43, 74, 82, 87),

Little work has been done on the reduction of /9-alkoxypro- pionitrlles. Ihitraore and co-woi%ers (87) hydrogenated di- (2-cyanoethyl)-ether and obtained a 25^ yield of a compouad

which they assimed to be di-(3-aminopropyl)-ether. It was subsequently shown by Wiedeman and Montgomery (88) that the coiapound reported by Whitniore, et , was very likely 3-

aalnopropanol. By carrying out the reduction of di-(2-cyano- ethyl)-ether in the presence of anhydrous asimonia the latter workers were able to isolate 20^ of di-(3-aminopropyl)-ether 36

and 35-40^ of 3-sanliiopropanol. Utermohlen (82) prepared nine simple 3 -alkoxypropylaiaines toy hydrogenation of the corresponding ys-alkoxyproplonltrlles in the presence of liquid ^monia and Raney nldfcel. The yields varied frora 25% to 78^, fhe formation of aecondaa^ amine was apparently responsible for some of the lower yields (around 50^)« ys-Benzyloxyproplonltrlle gave the expected amine, 3- benzylo:!^ropylaajine, in only 25^ yield. In this oaee con­ siderable cleavage was reported to have taken place at t^e oxygen atoia. This is not siiiprising inasmuch as benzyl ethers in general are readily cleaved by hydrogen over Raney niolcel to yield toluene and an alcohol (83). Purtheiroore, the forma­ tion of 3-aiBinopropanol in the reduction of dl~(2-cyanoethyl)- ether, as reported by Wiedemann and Montgomery (88), can only be accounted for by a slnjllar cleavage. It would appear, then, that the hydrogenation of yS-silkoxypropionltrlles may be com­ plicated by rupture of the ether linkage in addition to the usual side reactions.

Reduction of l,4-Di-(2-oyanoethoa^)-pentane

In the 500 ml. hydrogenation boiab there was placed 70 g. (0,33 sole) of l,4-di-(2-cyanoethoxy)-pentane and 4 g. of Haney nicikel catalyst. The borab was cooled to -50° In a bath of dry ice and acetone (in order to prevent excessive 37

loss of aiaiBonla by evaporation); 30 rol. of liquid mmonXe. was added and the bomb was sealed. Hydrogen was introduced to an Initial pressure of 200 atoospheres. Reduction began at o o about 80 , proceeded rapidly at 115-120 , and was coiaplete in o 50 Kinutes, The maxJjitBB temperature reached was 122 » The boab was emptied and rinsed with alcohol. Distillation of the coffiblned pix>duot and rinsings, after removal of the catalyst, gave §5,5 g, (77^) of l,4-di-(3-aminoprop03cy|-pentane boiling at 132-135° (g mo.),

d^j 0.9448, 1,4596.

Anal. Galcd. for 109.2j N, 12.8.

Pouadi H. S., 110.4, 110.0| H, 12.6. o A forerun of approximately 5 g., boiling over a 20 range, was obtained. The residue deooiapoeed extensively at 200-230^ whm an. attempt was made to distill It under a pressure of 1 mm. Another experiment using hydrogen at an initial pressure of 130 atasospheres gave substantially the saae yield of aisine (82,^). l,4-Di-(3-aralnopropoxy)~pentane is a colorless liquid which exhibits only faintly the odor which is-characteristic of aoines. It is soluble In organic solvents and sisclble with water in all proportions. Its aqueous solution is strongly basic to phenolphthaleln {a 5^ solution has a pH in excess of 11) and can be titrated to a sharp end point using 38 aethyl red indicator. Upon being exposed to the air, the ©sine rapidly absorbs carbon dioxide and eventusCLly foms a sticky gus« An atterspt to prepare the hydrochloride of this aalne by precipitation from anhydrous ether solution with diy hydrogen chloride gave an oil which oould not be induced to crystallize. The plcrate was likewise obtained as an oil* In view of the rather interesting properties of 1,4-dl- (3»amlnopropo3!:y)-pentane, it was felt that its fatty aold salts Bight prove to be of some value as surface active agents. Accordingly, the preparation of a series of these salts was undertaken. Dibasic acids were included, because the dljaaiii# salts of dibasic acids are important as nylon intersedlates (15).

Salts of 1,4-Dl~ (3-aminopropoxy )-pentane

Procedure

This salts were prepared in 2-3 gram quantities by the following general laethodsi

A. Solutions of the calculated quantities of the dlaslne and acid in appropriate amounts of absolute ethanol were well ffllxed and evaporated to di^ess. B. fhe calculated quantity of the solid acid was added to the diamine which had been dissolved in a little ©thanol# fhe solution was evaporated to diyaess* 39

C. The calculated amount of the solid, acid was added to a solution of tSie dlaalne In water. The resulting mlxttire was evaporated to dryness,

D. Aqueous solutions of the theoretical quantities of the diamine and aoid were well mixed and then evaporated to dryness.

Further details are given in connection witfo the Indi­ vidual es^erloenta.

Preparation of the salts

1.4--Di~(5''aainopropoxy)~pentane disulfaaate. Using isetli^d B, 2.22 g. (0,0102 mole) of l,4-di-{3-aiainopTOpoxy)-pentaji« and 1,94 g. (0.02 sole) of sulf^ic aoid gave a colorless oil which crystallized to an extremely hygrosooplo, whit®, waxy solid upon standing in a vacuum desiccator. The melting point o was about 65 , When an att^pt was made to reorystallize it fiTOB! a Blxture of absolute ethanol and ether, the substance darkened and melted at a much lower temperature,

1« 4~Dl-(5^aainQprQpoxy)-pentane diacetate. Treated bf method A, 1.20 g, (0.02 raole) of acetic acid and 2.IS g. (0.01 BJole) of l,4-dl-(3-a®lnopropoxy)-pentane gave a sticky oil which could not be induced to crystallize even after long standing in a vaeui® desiccator.

1.4-Pl"'(5~aiainopropoxy)"Pentane dip^plonate. linen 1.48 g, (0,02 mole) of propionic acid and 2.18 g. (0,01 mole) of 40

1,4-di-(3-anjlnopropoxy)-pentane were allowed to rerct In easentlal accordance ?;lth procedure A, th.ere rea'?lted a stj^clty oil 'w-ilch could not be brought to crystallization even -s.fter lon';^ standlpp; in a vacuum desiccator over concentrated sul­ furic acid. 1,4-Dl-(5~aiT;lnopropo>:y)-pentane dlbutyrate. In accordance

•with procedure A, 2.18 !•% (O.Ol mole) of 1,4-di-(S-r-mino- propoxy)-pent8r.e tre-'-ted with 1.76 g. (O.OS raole) of butyric acid. The result was a sticky oil which was placed in a vacuira desiccator. It resisted all attempts to cause it to solidify. 1« 4-Di-(3-&glnopropoxy)-pentane dilevulinate. By method

A, 2.32 g. (0.02 irole) of distilled levulinic aoid and 2.18 f". (o.Ol mole) of 1,4-dl-(3-anilnopropo3cy)-pentane gave a

sticky F?uia which could not be induced to crystallize. It l>ecerae dark brown in color on standing in a vaci'iia desiccator. A similar phenomenon has been observed in attempts to pre­

pare amnsonliaa levulinate by evaporating an aronsoniuo hydroxide solution of levulinic acid. 1,4-Dl-(3-aialnopropoxy)-pentan e dibenzoate. Benzoic

acid (2.44 g., 0.02 mole) ond 2.18 g. (O.Ol mole) of 1,4-

Ql-(3-aE!lnopropoxy)-pentane, •when treated by method A, gave a giffii. After long stsiiding In a vacuiaa desiccator, a few

crystals fonaed. In an attestfpt to cause the entire mass to

solidify, the mixture was triturated with a little ether and 41 absolute ethanol. fhe crystals dissolved satd an uncryst^- lizalile glass resulted. 1« 4->Di~(5*-aiainopropo3:y)»pentane o^calate* When treated by pracedure A, 1.26 g. (0.01 sole) of oxalic acid diii^drate and 2.18 g. (0,01 mole) of l,4-di*(9~aBinopropoxy)-pentaae gave m imiediate white precipitate, fhis was filtered, washed with absolute ethanol and dried to yield 3.3 g. of white powder raelting at 197-199'^. Reorystalliaation trm dilute ethanol gave white crystals of melting point 196-198°.

This salt was the only one of the entire series which was insoluble in absolute ethanol*

Anal. Galcd. for %3H28%^2* PouadJ 9*65. 1.4~Di'»(S-aainopropoxy )«>pentane aalonate* Malonlc acid

(1*04 g., 0.01 sole) and 2.18 g* (O.Ol aole) of 1,4-di- (3-aainopropoxy)-pentane gave, by procedure A, a colorless^ sticky gtffi which could not be caused, to ci^stalllze even on long standing in a vacui® desiccator* 1.4-Di~(5~8aiinopropQxy )-pentane succinate, in accordance with procedure A, 1,18 g. (O.Ol sole) of succinic acid and 2.18 g. (O.Ol aole) of l,4-di-(3-ai9inopropoxy)-pentsn6 gave a colorless oil which slowly ci^stalllzed to a ^it# ^lid oelting at 119-122°. Recrystallization fro® ether-absolute o ethanol gave a white powder of melting point 121-123 . Anal. Galcd. for 8.33. Poiaid: H, 8.42, 42

1.4-01- C5*affiiR0prQP0x?)'"petttane gXutarate* By B©tJ»d

B, 2.18 g. (0.01 mole) of l,4-di-.(3-ai8lnopropoxy)-peiitane ana 1.32 g. {O.Ol sole) of gl«tapic acid gavd a gnaa. fhls was alleged to stand in a vaeuiai desicoator with oceaaional stirring. It did not oi^stiglliEe. 1.4-'Di»(S-»aBinopropoxy)«»pentaiie adipate> ^en treated by pi^oedwre A, 1.46 g. (0.01 laole) of adipio aoid and 2.18 g. (0.01 ®ole) of l,4-di-'(3«a3slnopropoxy)-peats^e gave a clear glass. After long standing in a va&um desiccator substance began to orystalliz®. This was hastened by txd- turation with an ethe>-0thanol aiatture. The result was o a hard, white solid selting at 1S6»1^ . It was reo^stal- lized from ether-absolute ethsaiol to yield a white powder of tsi changed aelting point.

Anal. Calcd. for Cx7%6V2" Founds M, 7.92, 1.^-Pi"(5-gffiinopropoxy>~pentane sebacate. In aooord^c® with method A, 2.02 g. (0.01 sole) of sebaoic aoid and 2.18 g. (O.Ol sole) of 1^4-di-(3»aainopropoxy)-pentan# gave a glass which solidified to a light yellow, ciT'stalline sass on long standing in a Taotiua desiooator. It was extr€®ely hygi^s<^pio and ©elted indefinitely at about 60^, Upon re- ci^stallisatlon fro® a mixture of ether and abfN5luto etfeanol, the ooffipowd separated as an oil and could not subsequently be obtained in the solid state. 45

(S-aminepropoxy)«-peHtan& iE^eat»« l,4->Di-^3« a»lRop,mpo3cy)-p©iitane CS.IS g,, 0.01 mole) and 1,16 g, (O.Oi sole) of raalelo acid gsv©, "by method B, a grai which rapidly solidified to a white substance of melting point 120-122®. Beorystalliaation trcm a aixtiire of absolute o #thaiiol and ethep yielded a white powderjaelting at 129«1SI.,

Galed. for %5830^6% • 8,38. Pound 5 H, 8.6S»

1« 4-Di-»(S-aainoprepoxy) -pent^e itaeonate* freatseat of 2,18 g. (0,01 fflole) of l,4-di«(3«-8minopropoxy)«p6iitaii» and 1,30 g. {0,01 sole) of itaeonlc acid by prooedai^e 0, gave a gu® which resisted all efforts to cawse it to oxr^tal- lize# 1,4«-l>i»(S^^iaopropoxy )«-peiitane phthalate* Being sethod B, 2,18 g, {0,01 mole) of l,4~dl-{3-sal!iopropoxy)-'

pentan® was treated with 1,66 g, (0,01 sole) of phthalle acid. There resulted a crea® colored solid welting at 120- o 122 , After reoiystallizatlon fro® an ether-absolute ethanol Mixture, a white powder of melting point 125-126® was obtained, Mai, Cal«sd. for Found? H, 7,38,

1,4-M-(3-aainopropo3tY >-peatane dioxalate. Procedure

D was followed in the treatment of 1,09 g, (0»005 sole) of l,4-.di-C3-a®inoprc^xy)-pentane wit^ 1,39 g, <0-011 aole)

©f oxalic aeid dihydrate* fhe result was a white solid of o melting point 86*95 , Two i^crystallizations from aixtores of ether and absolute eth^ol gave a white powder ©elting at 44

108-110^. fhls oofflpeimd, in contraet with.the noOTtal sxalat*, was easily Boluhle in hot absolute ethanol# ^fi8l» Osilcdo foi* ?«04i» P^ufidli Rjj ?»20»

1.4-Pl*'C 5*aalBOprQpoxy)-peatane diaalonatf# Malonie acid <2.08 g,, O.0g aiole) and l,4-di-{3-'^inof>rop0xy)-peati»» (2.18 g., 0.01 sole) gave, method C, a. colorless glass i?hioh eoid.d not be iaduoed to cryetalllae, 1.4^P1~(g*aalttopropoxy )~peatane dianooiaate. tn &c

Mai. Oalcd. for C^gHggO^^QMg: H, 6.17. Foimd; H, 6.33.

1.4«'gi-(5»aaiziepropo:gy) -pentane dlarlutarate. When aethod 0 was applied to 2.18 g. (0.01 oole) of 1^4-di-(3- affiinop3PDpoxy)-pentane and 2.64 g. (0»02 sole) of glutario aaid, there resiilted a eolorless, sti^y gm whicfa could mt be bi^ught to ofyetaliization. 1 > 4-'l)i^(S-^emiMQtiTovQXsr)-peatane diadipata» IDs aoeerdance with pr©eedure 0, 1.09 g, (0.005 aiole) of l,4-dl»(3-sjBlB0- pi^p0xy)-pent8ne and 1.46 g. (O.Ol oole) of adipio aoid gave an oil. I^en allowed to stand in a vaouua desiccator this 45

slowly soliaified in n^settes of gllstenli^ neeaies of seltlng o point 97-99 • Raos^otalllzation fiHss a Blxtu3?e of ether and o sbBolute ©ths^ol gave white crystals melting at 1O2-10S •

Anal> Salod. for Cg3H4@03^o8gJ K, 5.49. FowadJ H, g.53. 1,4*51" (3*aaiinoprepoxy > -pentane diaaleaty* Applioation of method S to 1.09 g. (0.005 mole) of l,4-dl-C3-aainopropoxy)- pehtane and 1.16 g. (0.01 mole) of laaleie aoid gave an oil which# on standing ovemi^t in a r&&mm deslooator crystal­ lized to a white solid of meltii^ point 80-S5^* ^ea an att^pt was ®ade to reeiystalllze it froa ether-absolute ethenol, an oil separated from ttee eooled solutimi, and the compound could not be obtained in the solid state^ thereafter*

1.4>Di-' C inopi^pQxy)"pentane diphthalat^* Phthall® sold (1*66 g*, 0*0i sole) and 1*09 g* (0*005 sole) of 1,4- di-(3»a®inopropoxy)-pentane gave, by procedure B, a trans­ parent gm which resisted attwupts to cause it to oi^etalliEe* All of these salts were deliquescent substejicea which were very soluble in water. The resulting ®)lutions showed a strong ^deney to fosa* Wi^ the exception of the norsal oxalate, they were easily soluble in absolute ethsnol* fhef were^ of course, in«>liible in such solrmts as s'ttier ®ad ben2«a«. the sajority of the K»lid ooiapousds appeared to be iiiorphous, X ray exasinatioa* of the oxalate and succinate showed sharply defined powder diagr^s,

* Kindly carried out by Mr. Eferrey Dube. 46

Table III

Melting points of some salts of l,4-dl~(3-8iBinopropoxy)-pen tane

?r- Salt Melting point, C,

Disulfamate ca» 65 Dlacetate oil Dipropionate oil Dibutyrate oil Dilevulinate gim Dlbenzoate gum Oxalate 196-198 Malonate cum Succinate 121-122 ulutarate (jum Adipate 136-138 Sebacat© ca. 50 Maleate 129-131 Itaconate Phthalate 125-126 Dioxalate 103-110 Dlmalonate gum Disuccinate 92-94 Dlglutaratd gum Dladipate 102-103 Dimaleate 80-85 Diphthalate gua 47 indicating that the derivatives were microcirstalline in nature. The physical properties of the salts are smsarized in Table III. It will be noted that well-defined, solid salts were obtained only from dibasic acids having an even number of carbon atoms. 48

HYDROLYSIS CF 1,4-DI-(E-CYANQ£THOXY)-PfiNTATIE Mm RELAflSD a01!P0UMD3

Review of Methods by Which /3-A3.feox5n-jroplonlc AciiSs Have Been Prepared

Trie laajorlty of the previously renorted simple ;3-alkoxy- px'opionic acids have been prepared by the hydrolysis of their esters (29, 30, 36, 37, 61, 65, 65). The necessary esters have usuelly been obt^^.ined by a metathesis between a sodiiim alkoxide and a y3-halogenoproplonlc ester (2, 23, E9, 30, 36, 37, 38, 47, 61, 62, 63, 54, 55). ROf],, + XCH_CH„GOOR ^RQCHoGHpGOOR + NaX <' ad c. In practice, only alkoxlcles and esters having the eaae alkyl groups have been used, preeimiably in order to prevent the fon5£'t.1on of mixtures by alcoholysis.

An alternative and more convenient netliod for the ayn~ thesis of yS-allcoxypropionetes wo-'Id appear to be the addition of an alcohol to an alkyl acr:/late in the presence of the eorresponding: alkoxide. ROH t CHo; GHCOOR • >• RO CH,.OH., COOK». This reaction, which la analofTOUs to the addition of alcohols to acrylonitrlle (6, 7, 8, 17, 45, 51, 82), was first ob­ served by Purdle and Marshall (68) and Bubseauently reported by GoBs and Ingold (35) and Koelech (51). It was studied rather extensively in very recently published work (70). It seeras to be generally aviplicable to monohydric alcohols althoi^jh 49

In some cases the yields leave Bornething to be desired. An advantage of this method of preparation is the fact that the alkyl groups of the alcohol and acrylic ester need not be the BSBe (?0), A comparison of the cost of the y3-halogenopropionlc acida with that of the coBraeroial3y available alkyl acryla.tes further favors this reaction as a preparative method for y3-alkoxypropionic enters. An interesting and perhaps useful procedure was employed by JonoB and Powers (43) in the synthesis of y3-methoxypropionic acid. These '.workers added one raole of free ^-cl'iloropropionic acid to a solution prepared by dissclvinfr two grma atoms of sodium in excess raethanol and refluxed the resulting mixture. y6-Methoxypropionic acid was liberated fror- the dry sodium salt by treatnent with hydrochloric acid.

The literature contains a fe-? other references to methods by vriiichyS-alkoxypror;ionic acida have been prepared. These, althouf'h perhaps not g-enerslly applicable, raay be of Interest.

I'hey t?ill be mentioned here inasmuch as some of them will be referred to later. y3-Sth^xyproplonic acid hp® been j^repcred by the oxida­ tion of ^ethoxyproolonaldehyde ^"ith atmospheric oxygen in the presence of raanganous pcetete end ecetic acid (73) and by the hi^jh teiaperature reaction of diethyl ether with carbon dioxide In the presence of a catalyst (22). 50

^-n-propoxypropionic acid, and jS~JLso-propo3£ypropionic acid have been reported (58) as resulting froa the followinp; series of reactions: CsHr^OHgCHgOOCH;0(013)2 + O3

HpO ^ G3H7OGH2CH2COCOOH + {GH3)gC0

CgH^0CHgCHg(30C00H C3H7OGH2CH2OOOH + COg

Sokolow (77) and Wislicenus (90, 91) have obserred that when a ^ater solution of ^-iodoproplonio acid is heated with freshly prepared silver oxide, there results a mixture of yS-hydKJXypropionic acid, di-(2-carboxyethyl)-ether, and an unidentified isomer of the latter acid. From the mi^cture of the sodii® salts of these acids, Wislicenus (90, 31) extracted sodium p-i^droxyproplonate with hot 95^ ethanol. Ho then dis­ solved the sodium salt of di-(g~carboxyethyl)-ether in hot 90^ ethajiol and reported that this salt crystallized In long, white needles when the solution was cooled. Neither of these workers made any attempt to isolate dl-(2-carboxyethyl)-ether as the free acid.

While the presmit investigation was in progress a patent (9) was issued which desoribed the preparation of esters of di-(2-carboxyethyl)«ether. Di-(2-cyanoethyl)-ether was hydrolyzed by mineral acid and the resulting mixture was treated with an alcohol in the presence of an acidic catalyst. Here, too, the free di-(2-carboxyethyl)-ether was not isolated. It is worthy of note that none of these methods has been 51 applied to glyoolfi in an attempt to prepare dibfisic aoida of the ^^ <-alkylenedioxy-dipropionlc type. As a matter of fact. It seens doubtful that any of the classical methods would be entirely satisfactory for such a preparation. Except for the single reference (9) contained In the patent litera­ ture describing the preparation of esters of di-(2-carboxy- ethyD-ether, the ^-alkoxypropionitriles have apparently never been considered as starting materials for the synthesis of even the sinplesbp-alkoxyproplonic acids, Kilpi (48) studied the kinetics of the hydrolysis of ^methoxypropionitrile end ^-ethoxypropionitrile by hydrochloric acid In 50^ ethanol solution but did not report the products of the reaction* 52

Basic E^fdrolysls of y6-AIkoxypTOpionlti»iles

Basic hydrolysis of 1,4~di~(2»cyanoethoxy)'-pentaiie

I, To a solution of 12 (0.3 raole) of sodiua hydroxide in 100 ml, of water, there was added 21 g. (O.l mole) of 1,4- dl-(g-.oyaao0thoxy)-pentane. The mixture was stirred and hoated to gentle rafltix for 12 hours. Acidification and evaporation to dryness vacuo were followed hy chlorofor® extraction* Evaporation of thla extract gave 22 g# of li^t bro-5?n oil having the neutral equivalent 177, The calculated value for the expected dibasic acid is 124. The material showed no tendency to cjirstallize*

Distillation was attempted at a pressiare of 1 mm. ot less. Distillate was collected over a boiling range of 97® o to 200 * Mo definite frectiona were obsejrred and profound decomposition began #ien the temperature of the heating bath exceeded 135-140®*

II* This experiment was carried out in an attaapt to

purify the hydrolysis product as its ethyl ester. 'Bxe method chosen for the esterificatlon was one which had been successfully esployed by Drushel and Holden (24) in the preparation of esters of p-hydroxypropionlc acid. A ©Isture of 21 g. (0,1 raole) of l,4«dl-(2-cyanoethDxy)«

pentane and 10 g. (0,25 nole) of sodit® hydroxide in 100 si. 53 of wRter was refluxed gently for el^t hours. Forty silliters of 5.88 K Bulfurio add was added and the mixture was eiraporated.

The dried residue was thoroughly extracted with 150 ®1, of absolute ethanol. To the filtrate there was added 12.S g. (0,08 raole) of anhydrous copper sulfate and the mixture was gently boiled under reflux for 16 hours. After cooling and filtration, the etiianol was distilled from a water bath to yield a dark oil. This was sixed with water and extracted with ether, The ether solution was dried over sodli® sulfate and evaporated# The residue, which amounted to only el^t ollllliters, was distilled Iji vaouo. Pour grems of distillate O Q boiled fK>n} 100 to 165 (1-2 mm,). An equal quantity was tm- dlstlllable. The distillate gave a strong test for tl:^ presence of esters (20), but its further purification was not attempted.

Basic tedrolysis of dl-(2-cyanoethyl)^ether

To a 24$ (by ifeight) soditffli hydroxide solution (32 g,,

0.8 aiole, of aodliM hydroxide in 100 ml. of water) there was added 43.4 g. (0.35 sole) of di-(2-cyaiioethyl)-ether. The mixture was stirred and refluxed for seven hours and th®ci evaporated to dryness in vacuo. To the diy residue, 70 ml. of cold, concentrated hydrochloric acid was added. The 2?e- sultlng mixture was evaporated and extracted with chloi^fons.

Removal of the chlorofoiTii gave 51 g. of a viscous, tan syri^. The neutral equivalent was found to be 126. The calculated 54 value for the expected dlbas.lc acid, di--(2-oarboxyethyl)-ether, is 31• Caie f^ras of this acidic mixture was treated with thionyl chloride and then poured slowly into cold concentrated amiBoniuffi hydroxide. The resulting solution was evaporated to dryness and extracted \7ith dilorofom. Evaporation of the chloroform extract gave an imcryatallizabl® oil. Ho di-(2- carbamylethyD-ether, which has "been reported (14) to laelt at 146*^, could be isolated.

The remainder of the materisl was titrated to the phenol- phthalein end point with 3H sodium hydroxide s|id the solution was evaporated, fhe dry salt was extracted with boiling 95,^ ethanol. the residue was boiled ^ith 400 ml. of 90^ ethanol end filtered. There was a lai^e quantity of material which was insoluble in 90^ ethanol. None of the salt described by Wislicenus (90^ 91) separated froni the solution even on long standlJig In the refrigerator.

Basic hydrolysis of ^~ethoxypropionltrile

A solution of 36 g. (0.9 solo) of sodlti® hydroxide in 125 lal. \fater was mixed with 49.5 g. (0.5 mole) of )3-ethoxypropio- nitrlle and refluxed for four hours. It was acidified by adding 83 ml. of concentrated hgrdrochlorio acid and then evaporated to dryness vacuo. The residue was extracted with ether and the ether extract was distilled in vacuo. 55

Four greas boiled below 120° (20 mm,). Twenty graros of material distilled at 120-130® (20 mia.) with deooiapoBltlon

(heating bath at 180-220°). This fraction was not homogeneous and had a strong, vinegary odor. A consivierable quantity of light brom residue r^raained in the distilling fl.ask. fhe distillate dissolved in dilute aodium bicarbonete with evolu­ tion of carbon dioxide sjid reduced large quantities of 2^ potasx^iiffis pejaaejtganate solution with fonaetion of manganese dioxide. On standing ov©mip:ht the subat^ce polymerized to a transparent, colorless, elastic solid which was insoluble in aloohol or ohlorofoi© but dissolved slowly in sodiirat bi­ carbonate solution, irlth liberation of carbon dioxide, to

yield a clear solution of unusually high visooslty.

Acid Hydrolysis of ^-AUfcoxypropionltriles

Procedure

Hydrolysis studies were carried out by heating a mechani­

cally stirred nsixture of a /3-alkoxyproplonitrlle and one to

two tiraes the calculated qusmtlty of concentrated hydrochloric acid, fhe reaction tiroes and temperatures are given in con­ nection ^s-ita-i the individual experiments. After coispletlon of

the l^drolyals, the reaction alxtures were woi^ed up in accordance with one of the following procedures: 56

A. Th& reaction aixture waa evaporated to dr^ess in vaouo with heating on a water bath. The residue was taken up in acetone or ether end filtered with suction, fhB amroonitai chloride reaainlng was carefully washed wiiaa the solvent, llie liquid acids were isolated by vaouiaa distilleition of the oora~ blned filtrates. Th& solid acids orystallizied, of course, imon evaporation of the filtrates under diminished pressure.

B. The reaction mixture was diluted with water in suf­ ficient qiiantity to dissolve the precipitated amroonii® chloride. The organic layer was diluted with ether and separated and the aqueous solution was extracted several times witli ether. Distillation, in vaouo, of the oofflbined ether extracts gave tiie desired j3-alkoxyproplonic acids. The yleldd of the monobasic acids were calculated upon the basis of the neutral equivalents of the partially purified (once distilled) coapounds. With the exception of ^-ethoxy- proplonlc acid, they were then carefully purified by distilla­ tion, using a 20 m,, asbestos-wrapped, Vigreux fractionating coliMn, before the densities and indices of refrsctlon were deterrained. The yields of the dibasic acids were determined from the neutral equivalents of the crude coapounds. Neutral equiv^ent data plus agreement of reported and found values for physical constants were considered to afford sufficient identification for the liquid acids that had been 57

prepared pravioualy. Although the preparation of y3-n~foutoxy--

proplonio acid and yS-l^o-butoxyp'roploiilo acid Jms heen re­

ported (65), their physical constants do not see© to he available. They ^ere, therefor®, subjected to carbon and

hydrogen analysis. The monobasic acids were characterized, when possible,

by preparation of the solid o-^bromophenacyl esters in

accordance with a standard procedure (76). Another well ktiomi isethod (39, 76) served for the conversion of the dibasic acids to solid amides. Of the four amides obtained, three have

been prepared recently by other means (10, 11, 14),

Preparation of isoQobasio y^-allcorypropionic acids and their derivatives

fi-gthoxvpTOPionio acid. I. j6-Sthoxyproplonitrlle (49.6 g., 0,5 Bsole) was treated with 60 g. (o.6 mole) of con­

centrated hydrochloric acid for tm hours at room temperature and 45 minutes at OO-lOO*^. The escperlsent was completed by

method A to yield 16.5 Q, of unreaeted ^-ethoxypropionitrlle

boiling at 16?-171^ (ataaospheric pressure) and 22 g. of j3-ethox^ropionic acid of boiling point 11?-120^ (17 mm,) and 20 1.0635.

Anal. Galcd. for CgK^^^O^J N. S. 118.1. Foundj N. E., 119,7, 120.3.

When oorreoted for the amount of y3»ethoxyproplonltrlle 58

recovered, the yield was found to be 55% of the theoretical quantity. Palorasa (61) gives the constants foryS-ethoxyproplo- o 20 nitrlle as 'boilliig point 119-120 (19 rara.) and d ^ 1.0641. II, A Blxtui^ of 49.5 g. (0,5 mole) of y3-ethoxypropioni- trlle and 100 g. (1 mole) of concentrated hydrochloric acid was heated for six hours at 70-80° and 30 minutes at 90-100®>

®ie reaction fixture was wo2%:ed up by procedure A. fhe aoeton® solution was evaporated and distilled taider reduced pressure, fhere WS,B no fore-nm and there was obtained 50,5 g, (86^) of p-ethoxypvoplonlei acid of boiling point 118-121° (19 oia,),

P-SS^smllSlMSl y^ethoxvprooionate. fhe dei^ivatlve separated as an oil when the reaction mixture was ooolM. ®ie entire ©ixture was evaporated to yield an oil which slowly solidified, fhe solid was leached with water and dried, pi^jduct was reorystalllzed by dissolving it in a small quantity of 95,^ ethanol at room temperature and then cooling the solu­ tion to -20°, Lai^e, shining, white leaflets of melting point 47-48® resulted,

Calcd, for G^^H^gO^Br: Br, 25.4. Found i Br, 25.1

Ae a further clieclc on the identity of y3-ethoxyproplonio acid, two of its known derivatives were prepared. 59

Ethyl /^->ethoxyprot?lQnate» To a solution of 22 g. (0,186 mole) of ^-ethoxyproplonic acid In 35 ml. of absolute ethanol there was added 2 lal. of saturated, alcoholic hydrogen ciriloride, Th© lalzture was gently refluxed for 20 hours. Most of the excess alcohol was removed by distillation froia a water bath. The residue was diluted with an equal voltffle of ether, extre.cted three tlraes with saturated sodium bicarbonate solu» tlon and once with water. After drying over sodlun sulfate it was distilled to give 19.5 g. (72%) of colorless liquid o , , 20 of boiling point 65 <16 mra.), and d 0.9585. There ^as no 4 forerun and no residue. Palomaa and Kllpi (64) fTive boiling o . . 20 point 62-63 (13 laii.) and d ^ 0,9590 for ethyl ^etho^- propionate. /^-gthoxyprorjlonyl ohlorlde

37 g. (90^) of colorless liquid of boillner point 50-52.5° o PO (15 miBi) or 27-28 ( 2 mm.) and d ^ 1.0740. For^ethoxy- . . o proplonyl chloride Lelmu (53) reports boiling point 28-28.5 pf% (2 las.) and 1.0737. 60

p)-n-Propostyproplonlo acld» A mixture of 79.1 g. (0.7 Kole) of j0-n-propoxypropionltrlle end 140 g. (1,4 moles) of conoentrated hydx^chiorio acid ^ao heated for three hours at Q ® 70-80 and then for thirty minutes at 100 * tJslne acetone as the solvent, procedure A was followed to obtain 74 g. of colorless liquid "boiling: at 117-123° (13 mm.) and having tiie neutral equivalent 130. Thla represents a yield of 80^, Pure p'-n-^pvopox'ypropionXG acid has the following physical o on ^ oonstsntsj boiling point 87 (1 mm.), d'^ 1.0237 and xT 4 "" B 1.4233, Anal. Calcd. for 132,1, Founds M. E*, 132.0.

Physical oonatants reported (58) for p^n-propoxypro-' plonle acid are as follows: boiling point 110-112*^ (9 ibh.) and 1.4230, jJ •g-Broaophenaovl y^^-n-propoxypropionate. This derivative crystallized from the reaction mixture on cooling and had o the melting point 57-59 . It was recrystallized from a mixture of ethanol and water to yield glistening, -s^hite leaflets melting at 57-58'^.

Anal. Galod* for %4%704, Br: Br, 24.3. Pound: Br, 24.2. g~iso~Propox.Tproplonlo acid, Trestment of 79.1 g, (0,7 mole) of jS-lso-propoxyproplonitrlle and 140 g. (1,4 moles) of concentrated hydrochloric acid. In the sannpr described for the preparation of ^-n-propoxypropionlc acid, gave 55,5 g. of 61

o , . colorless liquid of boilinf* point 115-121 (13 mm.) and neutral equivalent 129. The yield tvas, therefore, 60^, fher© was a considerable quantity of brown residue. The physical constants for pure S»l^~propoxyproplonio acid ^ere found to o 20 20 he -boiling Doint 85.5-86 (1 raia.), ^ 1.0192, and n 1.4302. 4 "" D Anal. Calcd. for G0%2®3' 132.1. Found; N. £•,

131.3. Previously reported (58) pl^sioal constants for^-iso- o propoxypiKspionic acid are hoiling point 125-126 (22 rosi,) 20 and n 1.4202. ~ D p-Bromophenacyl ^-iso~propoxypropione te. The compound separated aa an oil from the reaction iai:rture. Evaporation of the entire mixture gave an oil which solida.fied on standing.

The solid was leached ulth water snd recryatellized by solution fit roora t®R!Dera.ture ^ siaall cjuantity of 96% ethsnol and o cooling to -20 . The result was tiny, white plates of melt­ ing point 44-44.5^. Mai. Calcd. for Br, 24.3. Poundt Br, 23.9.

-n-Butoxypropionic acid. A sixtiire of ^-n-butoxypropio- nitrile (76.2 g., 0.6 mole) and 90 g. (0.9 loole) of concentrated hydJTOChloric acid WES heated for three hours at 70-80^ end for o 10 minutes at 100 . Cotspletion of the experiment in accordance o with pi^pedure A gave 68.5 g. of crude acid boiling at 120-140

(13 ma.). The neutral equivalent of the material, 165, indi­ cated a yield of 69^. Pur® ^-n-butoxypropionic acid had the 62

following physical constants: boiling point 96-97.5^ (1 EH®,), 20 30 0.9929, and a 1.4268. ""4 "" D Mai. Galcd. for K, E., 146.1; C, 57.5; H, 9.67, Foundj M. E., 147.3; G, 57.1; H, 10.0.

t>-BroroophenaQyl y^-i^»butox.yp3?opionatg. This derivative crystallized from the reaction mixture on standing at room teroperature. Cooling and filtration gave product of m, p.

54-55°. !I^ils was recryfltall1 zed "by dissolving it In a little 95^ ethanol at jroom teraperattire and cooling the solution to o -20 . 01© compound consisted of shining, white leaflets roelt- Ing at 55^. •Anal. Calcd. for C^gH^^gO^Br: Br, 23.3. Poundj Br, 23.0.

-igOKButoxypropionic acid. ^-iso-Butoxyproplonitrlle (76.2 g,, 0.6 niole) was treated with 120 g. (1.2 nioles) of

concentrated hydrochloric acid for 3.75 hours at 75-8o'^. The rea,ctlon sixture was worked up by method B to give 63 g. of material boiling at 105-110'° ( 5 mm.) and having the neutrsQ.

equivalent 157. Hence, the yield was 67^€. The physical

constants for pure jso-butoxypropionic acid were found to be HO boiling point 89-90° (1 ma.), d . 0.9843, n 1.4227. - ^ ~D Anal. Calcd. for N. E., 146.1; C, 57.5;

H, 9.67. Found: N. E. 146.0; C, 57.0; H, 9.74. 63

^•~BrQBOph6naoyl /j-lBo-butoxyproplonate* The Aerlvativs oryetallized fz^om the reaction mixture on standing at room ^0 temperature as eoft, white, fibrous needles melting at 57-59 ,

Solution in a little dilute ethanol at room teraperaturo followed by cooling to -20*^ yielded white needles of melting point 57.5-50.5®.

Ansl* Calod. for 23.3. Found: Br, 23»2.

/^-se&.ButoxyproK>lQnio .aaia* A mixture of 76.2 g. (0,6

Bole) of /j-seo-butoxroropionitrlle and 60 g. (0.6 mole) of concentrated hydrochloric acid was heated for four hours at o 75-80 . Completion of the preparation by procedure B gave,

54 g* of crude acid havinij the boiling range 85-120° (5 bku. )* fhe neutral equlval(Mit, 161, indicated a yield of 56,^. Pure jS-seo-butoxypropionlc acid had the following phyeleal oonstantsj 0 f , 20 20 boiling point 90-91,4 (l mm*), d ^ 0*9946, and n ^ 1.4252* 4 "" J> Ans3.* Cslcdi for CijiH^j^^Og* M» , 146.Ij 0, 57.5j H, 9.67. Foimd; N. S*, 146*4; G, 57*3; H, 9*76.

Broaophenaoy1 /^seo-Butoxyproplonatei An oil separated frois the reaotion sixture on standing at rocMs tesperature. The entire mixture ms evaporated to dryness. 1?he resulting oil could not be induced to crystallize under any aondltlone* g-lBo-Amyloxyprooionlc g.oid« y6-leg-Aayloxypropionltrlle (84.6 g., 0*6 mole) ms treated with 120 g* (1*2 Eoles) of concentrated l^drochlorlc acid for 3*25 hours at 70-80°* Sub- sequait to completion ef the experiment in accordance with 64

pit)eeclure A, there wss obtained 76.5 g. of liquid of boil­ ing point 138-150° (14 mm.) and neutral equivalent 188, The yield was 69^. Twice fractionated fi-iso~atByloxypronionio acid had the follo'winff physical constants: boiling point 100*^ o , . prt 18 <1 mm,) or 13? (12 mm,), d*^" 0.9697, d 0.9725, and 20 ^ 1,4285. — p ^al. Calcd. for GgH^^gOg: H. E., 160.2, Found: W. E,, 162.7, Coisparable physical constants given in the literature o . for ;6-jj|o-aiayloxypropionle acid are boiling point 135 (12 18 mm,) (29, 30) and ^ 0,974 (36, 37)*

i>-BroBophenaGyl 5*lso'>ei3yloxyr)ropionate> This derivative crystallized from the reaction mixture on cooling to 0^* o Filtration gave fine, white needles melting at 53'-54 , fhe compound was reoiy atallized by dissolving it in dilute ethanol and cooling the solution to -20*', There resulted glistening o needles of oelting point 56 *

AngJ,. Calcd. for Cj^gHg^^O^Br; Br, 22.4, Found; Br, 22.2,

>8e<|»*.tovloxyoroplonio acid, A mixture of 70,4 g. Ig,5 mole) of yg-see-amyloxypropionitrlle and 100 g. (1 mole) of concentreted hydrochloric acid, heated for 3,5 hours at 75- o 85 and worked up by ®ethod A, gave 44 g, of crude product of boiling range 126-1^° (16 mm.) and neutral equivalent 179,

This corresponds to a yield of 49,2^, The low yield was du® 65

In part to accidental spillapje of come of th© ether solution before dlatillation. The pi^fslcal constants of the pur© o 20 cosDOund were boiling DOlnt 100-101.4 (1 mm.), a 0.9833, ^20 " ^ and a 1,4289, ~ D Anal, Oalcd. for N. S., 160,8; 0, 60.0; H, 10.1. Found: N. E., 159.4; G, 59.7; H, 10.3.

i3-g^,ff~,^y3,Qy,.yproplrQPs1^a' corapouna failed to crystslliae from the reaction niixtiire. ETaporation gaTe an oil which could not be caused to solidify under sjiy conditions, B-'Allyl&xypropionlc acid. A raixture of c. (0.6 wole) of ><3-.ailyloxypropionitrile and 90 g, (0.9 mole) of concentrated hydrochloric acid was heated at 70° for four hours. The experSsjent was completed by procedure B to yield 21 g, of lai- reaeted ^allyloxyproplonitrile cud 22 g. of crude y3-allyloxy- propionlc acid of boiling rsjige 93-120*^ (6 Effij. )* The neatral equivalent, 156, indicated a yield of 24^. IKiese two quanti­ ties of ffiaterlFtl ^fere fractionally distilled to yield 18*5 g. of ;0-allylo.^proplonltrile (boiling point 71-75^ at 6 roia.) o and pure ^-£llyloxyproplonlc acid of boiling point 111-112

(6 mm.) or 84^ (1 raia.), 1.0604 and § 1*4483. Coiv ^ D rectlon of the yield for the amount of nltrlle recovered gave a value of 33^*

Anal. Calcd. for OgH^oOg: N. E., 130.1; 0, 55.4; H, 7.75. Pouadi N. £., 130. a J C, 55,3; H, 3.17. m

P-B3?omopheaaeyl fi-AllyXoxyproplonata* fhe desired SOB- poiaid o^stsailzed from the reaotlcm slxture upoa cooling it o o t© . Filtration gave a white powder ®«lting at SS-3^ . It was reo;^«tallized hy dissolring it at roo® temperature ia a saall quantity of 95^ ethanol. fhe resulting solution was chilled to -20° and filtered to yield tiay, white plates of 0 seltlRg point 38-39 .

Anal* Galcd. for O^^H^gO^Brs Br, 24.4. Po«adJ Br, 24»3.

6*i2»MethQxyethoxy)«'PropioniQ acid. A solution of 77.4 g« Co*6 soie) of ^(2»»etho3grethoxy)-propionltrlle In 10(5 g.

(1*2 soles) of oonoentrated hydrooM-orlo aold was heated at 70-^° for three hours* fsing aostone as the eolvent, method

A was followed in worlcing up the reaction mixture. I^ere was o obtained 67 g* ©f produot boiling at about 12© (2 as*)* Sii# aeutral ©qui-ralentj 150, Indicated a yield of 75^* Pur® j3~<2-B0thaa3retho3sy)*proplonio acid had the following physieaJl 80 eonstaatsi boiling point 109*5-110.5 (0*S Mt*), d 1*1146 and a ^ 1*4356* Anal* Oalcd* for Gg^gO^J K. £•# 148*Ej C, 48*Sj K# 8*16* romds K. E*, 150*aj G, ^*3| », 8.3S* B*Brc^ot>henaoyl g-aethoxyethoxr )'>propionate. ®tils Qompouad s^arated f5w>® the reaction mixt^rs as aaersous, o white leaflets on cooling to 0 . It selted to an oil m waj^lng to roo» temperature and ooidd not be satlsfaetorlly 0 purified, fhe approxlsate melting point was 18 . 67

19ie p»&]LktkX^TQplonio aoida are colorless liquids harlng little ta» no @dop. They art ell soluble In the eoaiaon oz^anie 8ol•vents. In water, the ethoasy*, ieg^propoxy-, allyloss^-, asS

S-aetho3^thox3^» derivatives are easily soli^le wher«a# >^||r propos^ropionle aold is only slig^itly so. Mone of the hi^er ambers of the series is appreciably soluble in water. *Sh» pfc^sieal eonatants and yields of the oo^IlpG^«ds are given in fable I?, fable ? gives the oi^stalline foroi and selti^

points of the ^-broaophenaoyl jS-alkoxypi^pionates.

ISvalivef ^ ;3*rtl;p3!;ypyQpj,oi^l?

Di~(g'»oarboxyethyl)'"ether» A raixture of 63 g. (0.5 mole)

of dl-(S-cyafioethyl)-ether and 200 g, (2 soles) of conoentrateS l^rochloric acid was stirred vigoi^usly and heated for six o o hours at 70-80 and then for 30 aiinutes at 100 . ®ie reaction.

product was isolated by isethod A, ttsing aoetone as the solvesit* Evaporation of the extract gave SO. g, (100^5 of a tan syris^ whioh, upon standing overnight, solidified to a nearly white, crystalline mass, fhe neutral equivalent of this material was o 83«6 indicating a purity of It melted at 25-^ ®nid was rather l^grosooplo* It was very soluble in ether, alcohol or water, but Inmlv&ile in petrolet® ether. So®e difficulty was experifflsced in reozystallizing this eoropound because of its solubility characteristics and low

Beltii^ poin«» It had a strong t^ndimoy separate as an oil 58

Table IV

Fhysiosl oonstants and yields of y3-^koxyproplonlc acids HOCKgCHgCOOH

R Boiling point, 20 20 Yield, C# mm. B j3 i 4 %

Ethyl^ 117-120 17 1.0635 86 87 n-Propyl^ 1 80 120 13 1.4233 1.0237 0 IsO-Propyl 85.5-86 1 1.4202 1.0192 60 118 13 d n-Butyl 96-97.5 1 1.4268 0.9929 69 d Jjo-Butyl 89-90 1 1.4227 0.9843 67

sec-Butyl 90-91.4 1 1.4252 0.9946 56 e j8o-A®yl 100 1 1.4285 0.9697 69 137 12

sec-Arayl 100-101.4 1 1.4289 0.9833 49 2-Methoxyethyl 109.5-110,,5 0.5 1.4356 1.1146 75

84 1 Allyl 1.4423 1.0604 33 111-112 6 a ''Previously ' prepareil^ Faloroaa (61) gives boiling point 119-lgO® (19 mra.) knd ^20 i.0641. ~ A ^ Previously prepared. Llteratxire (58) gives boiliftg point 110-.112® (9 mm.) and 1,4230. " D ° Previously prepared. Literature (58) gives boiling point 125-126 (22 ms.) and n^O i.4gog. Previously prepared (65) but phyeioal oonstants not Imported. Q ® Previously prepared. Literature gives boiling point 135 (12 rnsa.) (29, 30) and 0.974 (36, 37). 69

fable V

Properties of p-brotsophenacyl ;6-alfcoxypropionates

ROCHgOHgOOOCHgCOGgH^Br-p

H Crystallise form* Melting point ®C.

Sthyl Large leaflets 47-48 n-Propyl Shiny leaflets 67-58 Iso-Propyl Tiny plates 44-44.5 n-Butyl Shiny leaflets 55 Jjo-Butyl Fibrous needles 57.5-58.8 seg-Butyl Oil

Jjo-Arayl Glistening needles 56 aeo-.toyl Oil

Allyl Tiny plates 38-39

2-Methoxyethyl Oil Ca. 15

# All of th© solid derivatives were white substances. 70 at te®peratiir®s far helom ite melting points !^e aoid ira« purified by dlssoliring it in hot ether, fo th« tsoiling solution, petiwletm ether {b. p. 60-70®) was added *mtil eloudinesfi Just appeared. fh& solution was elarified adding a few d3?ops of ether. It was then oooled to rotai teiaperature, seeded, cooled sloiily to a final t^peratur« of o aromd and quietly filtei^d. fhis was repeated ustil material of oonstant »elting point was obtained. di- {2-oarbo3cyethyl)-ether consists of hard> white oa?y8tal« aelt- o o , » ing at . fhe oompound boile at 189-19hg (1 m®* or lest} with slight de©<»}po8ition. toal. Oalcd* for OgHj^QOgt K* E*, 81.07j G, 44^ i| 6.18* rotadJ N. S. 81.77J C, 44.8; K, 6.56.

Pi-(3-oarbaayXethyl)»ether. A fixture of 1 g* (0*0062 ffiole) ©f dl-(2-oar^s;yet3iyl)-ether and 2 jsl* of thionyl chloride was refltixed gently for ^ sinutes* Tke exoesa thionyl chloride was removed by warning jy^ yaouo* fhe residue

m&B added dropwise, witto constant atirring, to 20 al* of oon- o oentyated frasoaii® hydroxide wiiloh was cooled to ->1© in & freezing mixture* Bie resulti% solution was evaporated to dryness taider reduced pressure and extraoted with l®t al^smts e^anol. Ivaporation of the extraot gave a light browa ®5lid o selting at about 130 . five reoryetallisatione ti?m be»sen€K absolute e^bmiol sixtures gatve eaall, glistiming, white niwmhB of constant selting point 143.6-144 . ©ae oc^pouad was easily 71

soluble in water or ethanol, "but insoluble in hm^z&n®, Anri. Galcd. for CgH^^gOgNg: K, 17,5. Foima: H, 17.4. Bruson and Riener (14) have reported the preparation of di-C8-»carbaffiyl©thyl)-ether by partial hydrolyels of the corresponding nitrlle with alkaline hydrogen peroxide. They o give the melting point as 146 . 1.(2-carl:^ryethoxY)-ethane, A solution of 65 g. (0.375 JBole) of l„2~di-(2-cyanoethoxy)-ethane in 75 g. (0,75 mole) of concentrated hydrochloric acid was stirred vigorously end heated at 70-80° for three hours. Then 76 g. (0.75 mole)

©ore liydrochloric acid added and stirring and heating eon- o o tlnued for 1.5 hours tt 70-80 and 30 minutes at 100 . fh© experiment was completed by method A. The acetone extract was evaporated to yield 77 g. of nearly colorless syrup. Upon standing overnight, this solidified to a whit© solid of melt- o ing point 4S-53 and neutral eaulvalent 110. The yield, there­ fore, was 94^,

©lis substance was soluble in acetone or \mter, slightly soluble in benzene, and insoluble in ether or petroleum ether. It had a pronoiaiced tendency to separate as an oil when re- crystallized. It was successfully purified by reoi^stalllza- tlon from benzene containing a little acetone* It was necessary to seed the slightly cooled solution and shake it oooasionally to prevent the separation of an oil. Five recrystallizatlons gave l,2-dl-(2-oarboxyethoxy)-ethane as spax^iling, ifhit© 72

o orystale of constant melting point 65 . Anal. Oslcd. for H. E., 103.1; C, 46.6; H, 6.86. Pound; H. E., 103.5; G, 47.0; H, 7.24. 1.2-Pi- C2"oarbaiaylethoyy)*ethane. A mixture of 10.3 g. (0.05 sole) of l,2-di(2-carboxyethoxy)-ethane and 23.8 g.

(0.2 mole), of thionyl chloriae was f:ently heated on a water bath for 1,5 hours. The cxcess thionyl chloride was removed by warslng jji vsouo. The residual oil was added dropwlse^ with constant stirrlni;;, to 160 rril. of concentrated amraonlua o hydroxide ffiaintaineA at -10 . Eva- or? tj.on of this solution in v&QxxG save & solid residue from 'vhich the amide was re- »fii imii I mn ' covered by extraction with hot ethsncl and subaequsat evapora­ tion. The crude oocpotind was recrystalllzed twice fs^m acetone end then four times from benzene-ethanol to f^ivs fine, white o Qvy'Jtala of conetant tseltino point 123 . The pure substanoe o underx?ent a sudden change in appearance at 107 on the rolcro o melting point block. Itoder rapid heating (10-15 per minute) o it melted rather shai^ly at 104 without resolldlfication. l,S-Dl-(2-oarba33yletlK>xy)-ethane is very soluble in water or ethanol, slightly soluble In acetone, but insoluble in hm- zene or ether. Anri. Oalcd. for Ga%Q04N2: N, 13.7. Found) N, 13.6. The preparatio'n of l,2-di-(2-carbaiBylethoxy)-ethane by partial hydrolysis of the oorresponding nitrile with 80^ o sulfuric aold at 55-60 has been reported (10). The melting 73

o point was given as 123-124 . 1.5-Pi"-(E-carboxyethoxy)-propane» A rolxtu3?e of 98 g.

(0,54 sole) of l,3-di-(2-cyanoethoxy)-propane and 216 g.

(2,16 moles) of concentrated hydrochlorio acid was heated at o ® 70-80 for four hours and at 95-100 for thirty fflintites. Procedure A was follot?ed in coiapleting the experiment. Evapora­

tion of the acetone solution under reduced pressure gave 118,5 g. of tan syrup which crystallized to a white solid feel ting point 65-77®) on standing overaight. The neutml equivalent

was 123, indicating a yield of 90^, The product was purified by recrystallization from ben­

zene containing a little petroleiM ether (boiling point 60- o. 70 ) to reduce the solubility. Seeding was necessary to pre­ vent the separation of an oil when the solution was cooled.

Five reci^etEllizaticnD gave pure l,3-di-(2-carboxyethoxy)- o propane whose melting point, 86-87 , was tmchanged on further recrystalllzation. The compound, consisting of soft ,whlte crystals, was easily soluble in water, alcohol and acetcaie,

fairly soluble in benzene, and insoluble in ether or petr^leiira ether. A dilute solution of the sodiuo salt gave no pre­

cipitate with solutions of the chlorides of aagnesim, barii® or calcitM. Anal. Calcd. for CgH^gOg; N. E,, 110,1; C, 49.1; H, 7.32. Found; N. S., 111,1; C, 49.3; H, 7.6E. 74

1.5~D1~(g~oarbainylethoxy )-PTOt)aiie. fhe pi^ooedur© followed was the same es that used in the preparation of l,2-di-(2-earb- aigylethoxy)-ethane, Prora 11»0 g. (0«05 raole) of crude 1,3-di- (S-carhoxyethoxi')-propOTe and 33.8 g. (0,2 mole) of thionyl chloride there waa obtained 13,5 r. of the crude amide. Tiiis uas recrystalllEsd fire tlses froia benzene-ethanol to give o fine, white cs^stals of constant laelting point 124 • fhe fflielting point of a mixture of this coiapound with l,2-di-C2- o cerbainylethoicy)-ethane was 95«-110 . l,3-Di-(2-oarb&Eiylethoxy)- propane is very soluble In water or ethcnol, slightly soluble

in acetone, and insoluble in benzene or ether.

Anal. Oalcd. for 0qEj^qO^2' 3.2,8. Found: H, 12.§• Bi~(3-72- carboxyethox'i7'-ethyl)-ether* A solution of 106 g* (0.5 moXe) of dl(2-/2-cyanoetho3gt7-®'^^^ in 2CX) g. (2 Boles) of concentrated hydrochloric acid was heated at o 70-80 for four hours after which the teiaperature was raised to 95-100° for 40 ainutes. The reaction fixture was woifeed

up in accordance with sethod A. Evaporation of "Qie acetone Bolution imder diminished preeeure gave 127 g. of tm eyrvsp

having the neutral equivalent 137, The yield, therefore* wes 91;t* Kiis oil could not be induced to crystallize under any

conditions. An att©npt to d3,stlll the acid at 0,5 i®. or o less failed when it began to decompose at about 2E5 • fhe derivitization of the compound as the knom, solid amide leaves no doubt as to its structure. 75

PI-(2'-/2-cai'baaylethoxa;/-'ethyl)~ether. fhe aaide was prepared in the usual way using 16 g. (0.064 mole) of crud© di~tg-/S-.carhoxyetlio3g27-®t^iy3.)»0ther and 30.4 g, (0»S56 sole) of thionyl chloride. Three reoiystslllzatloRS (two from aceton® and one fixisn ethanol-benzene) gave tiny, sparkling, whit© crystals of melting point 103-103.5 • Di-(2-^-carb~ aaiyletlio3^-ethyl)-.eth0r is vei'y soltible in mter or ethanol, fairly ^luble in aeetone» and insoluble in benzene or ether,

Anal. Calcd. for G^o^20®5^2* FormdJ H, 11.2.

f^iis compound has been prepared previously by partial hydrolysis of the corresponding nitrile with 80^ uulfurie acid at 55-60° (10), and by heating diethylene glycol with aoiylaaiide in the presence of trimethylbenzylaisaonitM hydroxide (11). The melting point reported in each case has been 103- 0 104 . The solubility characteristics of the dibasic acids of the y5-allcoxypropionie type have been given in connection with the Individual experiiaents* Ihe melting points of the eora- pounds are suisraarized in fable VI. The melting points of the amides^ which served as derivatives for the acids, are given

in Table VII with appropriate references* 78

Table VI

Melting points of )3-alkoxyproplonic acids of the type HOGOGHgCHg-Y- igCHgCCOH

Melting point, ®C.

-0- 60-61 -.OCHgOHgO- 56 -OGHgGHgCHgO- 86-37 -OCHgGHgOCHgCHgO- Oil

Table VII

Melting points of ^-alkoxypropionaiaiaee of the type

HgHCOCHgCSHg - X- CHgCHgGOMg

1 Melting point, ®G. Po\md Reported

144 146 (14) -OCHgGHgO^ 123* 123-124 <10) -"OCHgCHgCHgO— 124* -OOHgGHgOCHgOHg^ 103-103.5 103-104 (10,11)

*^1^e mixed melting point of these two substances was 95-110°. 77

Attespte to prepare 1.4-dl-(2'-carbox .Tetho3gy)-pentane

MM M1, a mixture of 40 g. (0.19 mole) of l,4-di-(2-c3ranoethoxj)-p6ntane and 76 g. (0.7S mole) of concentrated hydrochloric aoid was o stirred and heated under reflux for five hours at 80-90 and o for 15 minutes at 100 , A ^'hite solid began to precipitate after 20 minutes* heating. When water was added to dissolve the aiaroonliaa chloride an insoluble liquid layer appeared# Thda WES extracted with chloroform, fhe ohlorofor© extract i?ES evaporated and dried ^ vacuo over sulfuric acid to yield 40,5 g. of a light bii^wn syrup which formed a glass at -76° and could not be induced to crystallize under ^y conditions. The neutral equivalent was found to be 199 (Celcd, for

Cj^lHgQOg* K» S»J 124)« A solution of 38 g. of the material in 58 ml. of absolute ethanol to which had been added 3 ral, of saturated, absolute alcoholic hydrogen chloride was gently refluxed^or 24 hours. The excess ^oohol was distilled from a water bath* The

residue was diluted wilAi an equal voluae of ether, extracted twice with saturated sodit® bicarbonate solution and once with water. The ether solution was dried with soditai sulfate and distilled vaotio#

Three fractions were obtained. The first boiled at 85- o 96 (1 ra®.) and consisted of 17 g. of a colorless liquid

which contained chlorine, but no nitrogen and f:ave a positive 78

test for the presence of esters (20). The Interae^late o . . fraction of 7 g., boiling rouge 96-146 (1 mm.), \fas dle- o carded. The third fraction welded 9 g,, boiled at 146-175 (1 mm,), contained chlorine and nitrogen and gave a positive ester test. If, during th© ig'drolysls of l,4-dl(2-cyanoethoxy)-

peniane, one of the ether linkages were ruptured by the action of the concentrated hydrochloric acid, a ^S-Cchloroamyloxy)--

propionic acid might be one of the resulting fragraentB. fhis eoropoiaid m?uld appear ea the ethyl ester under the oonditlon® of the foregoing experlsent. ^e principal fraction obtained

above was purified and analyzed in view of the poasiblli-^

that it night be an ethyl p-(chloi^m8ylosy)-proplonatei Sub­

sequent to redistillation^ there V8.s obtained 13 g. of o P5 material of boiling point 109-112 (2.5 tam.), 1.0325, 25 and n 1*4586* D Anal. Calcd, for Mrp» 06.63; C, 53.9; H, 8.^; Ca, 15*9. Found: Mr^j, 56*75; G, 53.6; H, 8*82;

CI, 16.2* Alooholvsis of 1^4-di-(2-o7ano8thoxy)-pentane. I. The method of Spiegel (78) was used. A slxture of 22 g. (0.105 sole) of l,4-dl-(2-cyanoetho3ty)-p©ntane, 4.14 g. (0.9 mole) of absolute ethanol and 19.6 g. of concentrated sulfuric acid was heated under reflux in an oil bath for six trours. Upon

cooling the mixture and pouring It Into water, an oil 79 separated, fhls was extracted with ether, dried over sodiiaa sulfate and distilled at a pressure of 2 mm, iSie material boiled oTer the entire tsiperature range of 65-200*®. Dletll- lation was stopped when decoraposition began. Mo cleajj^eut fre.otlons were observed and there was considerable undlstlli-- able residue. n. fhe procedure succeasfully used by Sabetay (72) in

l^e alcoholysis of ajkoxyacetonitrlles was followed, A Bolutlon of 27 g. (0.74 laole) of dry hydragen chloride iM 46

g. of absolute ethanol woe prepared, fo It there was added

21 g, Co.l mole) of l,4-di~( 2-cyanoethoxy)-'pentane. fhe re­ action aijcture was protected from the aetion of atmospherio moisture by aeans of a ealolum chloride tube, fhe solution

became wara at once and after standing for tm hours a whit® precipitate had ftorraed. The aixture was tlim refluxed for three hours. The precipitated amsonlua chloride was filtered

and washed ^th absolute alcohol* WiBn dry it weighed 10 g. (Idieoretlcal qiiantlty 10,7 g.). The ethanol solution was

diluted with ether, extracted with eodiiai bicarbonate solution

and washed with water. It was dried with «odlm sulfate and o dlatllled. Four grams boiled at 65 (15 am.). The remainder boiled fro® 55° to 200^ (2 ram.) and no clesr-cut fractions

were observed, ©lere was approximfttely 0.5 g. of dark residue. 80

Preparation of 2~carbethoxyethoxy^-»Dentane. I.

The procedure used waa patterned after that of Kimball, Jefferson rnd Pike for the preparetion of ethyl ot-phenj?!- acetoacetate (49). A mixture of 75 ml. of absolute ethanol (distilled fro® sodluro and diethyl phthalate) and 21 g. (O.l laole) of l,4-dl-(2-oyanoethoxy)-pentane was saturated

with an^dj^us hydrogen chloride during five hours. o mixture was maintained at a temperature of -10 , stirred aechenloslly, and protected froni atmospherie moisture

means of a calcium chloride tube during this trealaaent. It

was then allowed to stand at room temperature for 24 hours. After rmoval of essoess hydrogen ciiloride gentle heating in vaouo. the solution was poured, with vigoTOus shaking,

into a mixture of 20 g. of nodlum carbonate, 120 ml. of water, and 200 ml. of ice, T^ils was extracted at once with three 50

ml. portions of ether, fhe ether extract wbm washed vlth foin? 50 ml. portions of ice-cold sodium chloride solution. A solution of 20 g. of concentrated sulfuric aoid and 140 ml. of water was shaken with 300 ml. of Ice until frost formed on the outside of the flask. 0ne»*ha3i.f of th5-s mixture was poured

into the ice-cold ether solution. After 12 seconds shaking the acid layer was separated. The remaining acid was added in two portions. After each addition, the mixture was shaken for 15 seconds and separated as before. The ether solution

was saved. The clear, colorless acid solution was heated on 81

o a water bath, for 30 sinutes at ^ . It soon beoaae olouay and an Ifflroiscible layer appeared. This ms separated, after eooling, and the aqueous solution was extracted with 25 ml. of ether. The ether extract we.e washed with 10 ml. of water which was added to the p.cid solution. I'he acid solution was o heated at 80-90 for 45 minutes, cooled, arrf. extracted as before. The corabined ether extracts were washed iJith 25 ol, of 5^ soditj® bicarbonate solution and 25 al, of water, and finally dried over anhydrous sod.ixaa sulfate.

Upon distillation under diminished pressure^ there was obtained 2.6 g. of material boiling at lOQ-151® (l ram.) and

10.4 g. (33^) of 1,4-di-(2-cErbethoxyethoxy)-pentane of boil­ ing point 161-153® (1 igtm.). The first fraction contained a trace of halogenj the second contained none, Beth fractions gave strongly positive tests for the presence of esters (20)*

Heither fraction contained nitragen. II. Sixty grms (0.15 mole) of coarsely pulverized bls-(ethyliminoester hydi*ochloride) of l,4-di-Cs-cyanoetho:ffy)- pentane wm added in small portions to 200 ml. of distilled water at room temperature. I^e addition was csoprled out over a period of tw€mty ainutes and the mixture tos stirred con­ tinuously. The solid dissolved readily and no teraperature change was observed, but an oil begeJi to separate at once.

After stirring for fifteen minutes at room temperature the fflixture was heated at 45° for thirty ainutes. It was th^ 82

oooled and extracted twice with 50 ml. portions of ether. The other extract was washed with 25 ml. of water which was returned to the reaction mixture. With continued stirring the mixture was heated at 90-95^ for 45 minutes. It was cooled and extracted with two 50 ml, portions of ether. The coEsblned ether extracts were trashed with 50 ml. of 5%

eodlm bicarbonate and with 50 ml. of water. After drying over sodii® sulfate the product was dietllled jUi vacuo. There was obtained 7.5 g. of material of boiling range o , . 60-142 (0.5 raiB.) which contained chlorine but no nitrogen, o The desired product was collected at 142-145 (0.5 aun,). It

weighed 28 g. {58^ yield).

l,4-®l-(2-0arbethoxyethoxy)-pentane Is a colorless 25 25 liquid of d , 1.0174 and a 1.4363. — 4 «• ])

Anal. Galcd. for Squivalent, 152.2; C, §9,2; H, 9.27. Foundt Saponification Equivalent, 151.4, 152.4; C, 59.5; H, S.54.

Att^pt#4 stX. I* pentane. The conversion of l,4-di-{2-carbet}iox3rethoxy)- pentane to the dlsodli® salt of l,4-di-{2-carboxyethoxy)-

-pentane, by quantitative saponification, was attempted. A sample of l,4-dl-(2-carbethoxyethoxy)-pentane weighing 1.028

g. was placed in a 25 ml. flask and 5 ®1. of 95% ethanol and

two drops of phenolphthalein were added. a buret, 1.29 M alcoholic sodiua hydroxide was added until the solution was 83 basic. Kie ralxture was allowed to stand at roo® teisperatur© wntll it beeaise colorless. It me again raade basic by the addition of a little of the standard base. These operations were continued. After about three hours, ^en approximately

50^ of the calculated amoxmt of base had been added, the mixture began to exhibit a strong odor of ethyl aorjlate. This sugfjested that the ester might have been deeomposing into 1,4~p®ntanediol and ethyl aorylate. llie experteent was abandoned. 84

DISCUSSION

©1© mechanise by which -^-valerolactone undergoes the

Friedel and Crafts inaction appears to be tee saaie as that for the simple esters. Evidence in support of this is the observation that little or no jfe^drogen chloride is evolved fTom the reaction mixture until the catalyst-lactone ratio

slightly exceeds unity. The first mole of catalyst probably brings about the formation of an aluminiaa salt of 7^-chloTO-

valeric acid which subsequently takes part in the alkylation reaction. Of incidental Importance is the fact that the reaction of -^-valeiHJlactone with alwalmm chloride or related com­

pounds ®lght afford a means for the synthesis of ^-chloro-

valeric acid whl<^ irould be more convenient ^an the sealed tube procedure of Wohlgemuth (92), y-Phenylvalerlc acid itself has been prepared during the course of a nusiber of studlea (50, 55, 57, 84), However,

except for the work of Sljkaan (25), It has Invariably been

made by laborious five or six step reactions. Sils has no doubt been due to the difficulty of obtaining -^-valeiolactone, because the advantage of the single operation of tl® Friedel and Grafts sethod of synthesis are obvious. An extension of the reaction to aromatic compoiands other than benzene and toluene should make readily available a ntasber of ^-arylvalerlc 85

acids. Any farther study of the use of T'-valerolaatone in

the Frledel and Crafts reaction should include.an investiga­ tion of catalysts other than aluminum chlorld©* fhe polyalkylatlon of benzene by means of -^-valerolactone

can be effected in good yield. However, the oonfln®ient of the reaction to dialkylation does not appear to be li®edlately

possible since. In all the experiments, there were obtained

ralxtures of mono-, dl-, and possibly tri-alfcylatloa products. On the other hand, a crude mixture of this sort, containing mono-, di-, and trl-basic acids (or their esters) mi^ht be worthy of investigation as an intermediate in the prepara­ tion of a ^viscosity-stable^ (34, p, 725) polyester or

polyaiaid©. On a production basis, the orientation of the substituents in the benzene ring would not be important. !Sie formation of 2-Bethyltetrahydrofuran in the hydro- genation of-jr-valerolactone over coppea?-chrc^iti® oxld® has

not been observed heretofore. It has been reported, however, that reduction of the lactone in the presence of Haney nickel gives soiae 2-Biethyltetrahydrofua?an (l, p. 78). In this case

It was assuaed that the compound resulted fpom direct hydro- genolysls of the carbonyl oxygen without mipture of the lactone ring. !Hie esse with which 1,4-pentanedlol can be dei^drated to fora the cyclic ether (52, 33, 56) suggests that in the copper-chrc^iuai oxide reduction the dlol Is the precursor of E-methyltetrahydrofuran. In support of this 86

aeohaiiiffls Is the fact that hi^er reduction t^aperaturefl result in lower yields of 1,4-pentanediol and increased rousts of 2-aethyltetrahydrofuran, The two-stag® reduction of ethyl l^vullnate can be esqplalned on the basis of the report that the oarhcnyl groiQ) is hydrogenated orer copper-chromliffl oxide at a tmperatui^ o of 1E5-1®3 (1, p, 51), whereas the hydrogenolysie of the o ester linkage takes place at 200-250 with the msm» catalyst (1, p, 97), It appears likely that the first step of the reaction Involves ttie reduction of the ketonic ester to fOTm ethyl y-hydroar^alerate. Hydrogenolysis of the latter cofflpoui^ could then take place at a higher teaperature wi'Qi

the foxiaation of 1,4-pentanediol and ethanol* Consideration of the fact that the esters of'^-hyd:mxy-

valerlc acid are reported to imdergo decomposition into

•^-valerolactone and an alcohol at ®oderately high teiapem- tures (81) suggests another poesihility for the second st^

of the reaction* Ethyl ^-hydroxyvalerate could decoajpoe#

into -^-valerolactone and ethanol. hydrogenolysis of th® lactone would th.en occur in the usual way. fhe end products would be the sajse in either case.

•Phe reduction of levulinic acid over copper-chroaim oxide at a teraperattire and pressure higher than those fisa- ployed by Bromi (4) gave a soaewhat higher yield of 1,4-

peatsnediol and a greatly deci^aaed mGxmt of T'^-valewjlactone. 87 fh© total recovery of products was not as good as that re­ ported "by Brown (4), however, and the relatively large quantity of low boiling material obtained si^gests the fonss- tlon of 2-®0thyltetrahydrofuran and mixtures of asyl alcohols. The use of mixtures of Raney nlcikel oopper-oferomltis oxide catalysts in attempts to effect the direct reductloa of levullnlc acid to 1,4-pentanediol might prove to be aa interesting and l^ruitful study. The fact that slightly lower yields of y3-alS:oxyproploni- triles were obtained from secondary alcohols is in accordance ^ith the observations and suggestions of UtersoKIen (82), In- fortsation gained during the preparation of ^-sgg^butoxy-

^~S®§-smyloxyproplonitrlles si^gests further that the heats

and rates of reacticsi may be lower In th© addition of seoondaiy alcohols to aci^lonitrlle.

It se^s noteworthy that in the hydrogenatlon of 1,4~ ai~(E-cyanoethoxy)-pentane at 120-125® little or no evidence for cleavage at the ether linkages was obtained. It will be recalled that Utenaohlen (82) reduced y3-allcoxyproplonitr5.1es o o at 90-100 and reported lowered yields at 125 as a result of oleavs^®. ^e very high yields of hydrogenolysis products

reported (88) to occur In the reduction of a related ooapound, di-(2-cyanoethyl)-ether, aay indicate that two S^oyanoethyl gftjups attached to a single oxygen atoa aake the syst«B particularly eusceptibls to rupture. 88

The excellent yield in which l,4-ai-(3-^inopropoxy)- pentan© ean "be prepared, the coaplete water-aolubllity of the ooapouisd, and its high basic strcmg'^ sake it extremely Interesting as a ssaterial for further general inirestigatlon. !Kie interesting fact that solid salts of l,4-di-{3- aisinopropoxyj-pentane were obtained only n^en t!te acids were dibasic and contained an even maaber of carbon atoms has been pointed out. Determination of the ci^etal structuree of the solid salts by means of X ray studies si^t suggest

an explanation for this phenomenon. fhe complete failure of the attempts to saponify

y3-alS:oxyproploaltrlles by treatiaent with 9-g2jg sodiua hydroxide solution was unexpected. Under the same conditions Bruson 0«d Mener (12, 13) obtained excellent yields of acids from the 2-cyanoethyl derivatives of active aethylene eoEi- pouads. In these cases, however, the S-cyanoethyl groi;Q)S were attached to carbon atoms,

fhe properties of the Material isolated fro® the basic hydrolysis of ^~etho3cypropionitrile si^gested that it might have been aos^lic acid. A consideration of the facts that the addition of alcohols to ao3?ylonitrile is catalyzed by bases and that an equilibrlii® is reached bwtween a y5-alkoxy- propionltrlle on the one hand, and aorylonitrlle and m alcohol on the other, say throw soae light on the problai. The addition of a base to a pure ;6-alkoxypropionitril® aight 89

be esspected to cause liie formation of some acirlonitrile and an alcohol. Under the conditions of basic hydrolysis it would seeffl that th© followinf^ reactions could take place:

ROCHgCHgCM > RCMi + CH2 ^ CHGH

R0GRgGH2CH + NaOH + HgO ^ ROGHgCHgCOOHa -»• KH3

GHgjGHGM + HaOH HgO—CKg-.CHGOOSa + HHg

If it is aseuaed that the conjugated syst^ of double bonds which takee part in the Michael condensation ia modi- fled in sodixai aorylate, then this cofflpound woiild mt be

expected to add an alcohol, fhe relative asoimts of the

various substances ultlaately resulting would depend upon the rates of the competing reactions lu-rolved. It se^s poasible that this ©echanisra could account for the foroation

of the cofflplex sixtures obtained in the basic hydrolysis of p -alkoxypropionitrilea# fhe acid hyd2:^lysis of yS-alkoxypropionitriles seeas to be a fairly general method for the preparation of ;0-alkoxy- propionic acids. Slightly lower yields were obtained in the saponification of the g-seo-alkoxypropionltriles and there was a general decrease in yield as the size of the al&yl group increased. Excellent yields were obtained wi-tti the four bifunctlonal ;0-alkoxypropionitrile8 containing only

prlaary ether linkages, Inassttch as little attempt was aade to detersiae 90 optiau® conditions for the hydrolysis ofjS-alkoxyproploni- triles, it is quite possible that the procedure may be sus­ ceptible to improv^ent. fhe use of dilute acid and longer reaction ttoes sight isproTe some of the yields. Meyerthe- lesa, it is felt that the successful hydrolysis of ^-alkoxy- propionitrlles affords a new and conyenient method for th® synthesis of aany ^-alkoxypropionic acids. It should be

f partictdarly valuable In the preparetion of ttie diojqfdiproplonio acids for which it is doubtful that any of the previous metiKJds would be satisfactory. It seems possible that certain coapounds of these types aight find application as wetting or flotation agents.

It is difficult to understand the deooiapositlon of l,'^-' di-{2-oyanoethoxy)-pentane upon hydrolysis with hydrochloric acid, fhe fact that lower yields were obtained in the saponi­ fication of JG-see-alkoxypropionltrlles suggests that the COIB- pounds containing secondary ether linkages may be aore sensitive to concentrated hydrochloric acid. Should this be

the case, then l,4-dl-(2-cyanoethoxy)-pentane, u^on hydrolysis

with concentrated hydrochloric acid, may underijo a cleavage of soBie sort at the secondary ether liiAag© with subsequent foraiatlon of a mixture of hydrolysis products, fhe isolation of a chlorine- containing compound from the reaction aizture indicates a rupture of the molecule with addition of tii«

elements of hydrogen chloride. The tentative identification. 91

based upon analytical date, of this corapouncl as the ethyl estor of a (chloroamylos:y)-proplonlc acid is certainly in support of this sur^p:estion.

The successfvO. conversion, at lov? temperatures, of 1,4- di-(2-c7anoGtho3cy)-pentane to the ethyl ester of the cor- respondinp- acid, indicates thrt the oompotrnds ar® stable under sufficiently nild conditions. However, the low- yields obt.'-'lned, and the length and complexity of the pre- pp.rrtioD, me^e: questionable the value of further investiga­ tion of this problem as an approach to the utlliEation of

-^-vslerolEctone and 1,4-pentanediol, 92

SiaJMAHSC

1. fhe preparation of y-phenylvaleric acid from bsniene and'^valerolaetone by means of the Frledel and Crafts re-

aetlon ms re-exaoined. The -^phenyl^alerio acid thus ob­ tained was shown to be identical with that prepared by other aethods, Ftcm -^phenylvaleric acid, th« solid p-bromo- phenaeyl ester was prepared to furnish a derivative for

futtire reference. 2. It was found possible to cause benz®ne to condense

with more than one molecular equivalent of -valeTOlaetone to yield a mixture fro® which ^ ^-phenylene-Clivaleric acid was

isolated as the diethyl ester. The position of substitution in the benzene ring of this compound was not detei®iasd» 3. Some previously described methods for the prepaj®-

tion of 1,4-pentEnediol were reinvestigated. In each case ii^roved yields of 1,4-pentanedlol were obtained. 4. 2-HethyltetrahydrofuraR was isolated and Identified

as a by-product In tiie hydrogenation of-^valerolactone to 1,4-pentanediol. 5. The addition of acrylonitrlle to 1,4-pentanediol gave l,4-dl-(2-cyanoethoxy)~pentane which 'sms characterized by conversion to the solid bis-Cethyllainoester hydi^ohloride),

6. A series of sew yQ-allcoxyproplonltrlles was prepared. 93

7» l,4-Dl-(2-cysnoethoxy)-pentane was hydrogenated to yield l,4-dl~{3-sffilnopropoxy)-pentaui0. Proia this divine a series of 3s,lts was prepared using sonobasio and dibasic organic acids. Solid salts were obtained only fro® dibasic acids oontalnlng en even number of carbon atoms.

S. Basic hydrolysis was found to be imeatisfactoi^ for

the conversion of y3»ethoxypropionitrile, di-(2-oyanoethyl)~ ether, and l,4-dl-{2-cyanoethoxy)-p0ntane to the corresponding acids.

9, By means of acid hydrolysis, a series of new aono- basic and dibasic acids of the y3-alkoxyproplonic type was prepared, fhese were characterized when possible by suitable solid derivatives such as p-bromophenacyl esters or simple asssldes#

10* l,4-I)i-(2-oyanoethoxy)-pent8ne, when subjected to

acid hydrolysis, underwent an obscure decOTtposltlon. Frcm the resulting aixture, a chlorine-containing acid was isolated as the ethyl ester* The analysis of this ester was in close

agreeeent with that calculated for an ethyl ^(chloro^yloxy)-^ proplonate4

11. Tm laethods were found for the conversion of 1,4- di-C2~cyanoethos:y)--p©ntane to l,4-di(2-earbethoxyethoxy)- pentane, the ethyl ester of the corresponding acid. Both of these methods involved the fonaati^n, at low temperatures,

and the hydrolysis, at ordinal^ temperatures, of a sinei^l acid salt of the Ms-ethyllsinoester. 94

ACaCNOWLEDGEMi^TS

!Hie author wishes to e:i^rese his sincere sppreolation to Dr. R. M. Hlxon for his enoouragC' aent and advice given throughout thle research,

fhe advice of Mr. D. S. Flikkesta was ex- tr^ely helpful in the slcroaiialytioal wos^ carried out during the investigation. 95

LITERATURE CITED

1. Adklns, ^Reactions of Hytlrogen with Organic Gonspoimds over Copper-Chroroliaa Oxide and Nickel Catalysts, '* The University of Wisconsin Press, Madison, Wisconsin, 1937, p. IS. 2. Bouveault aBd Blanc, Bull, soc. ohiro, 737. 31. 1211 (1904), 3. Bowden, J. Am. Ghero* Joe., 60, 645 (1938). 4. Brown, H. D, Unpublished studies. lowct State College. 1943. 5. Bruson, J, te. Ghea. Soc., 64, 2457 (1942).

6. Bruson, U. S. Patent 2,280,790, April 28, 1942. Original not seen. Abstracted in jg. A., 5588 (1942). 7. Bruson, U. g. Patent 2,280,791, April 28, 1942, Original not seen. Abstracted in C, A., 36, 5589 (1942).

8. Bruson, U. 3. Patent 2,280,792, April 28, 1942. Original not seen. Abstracted in A., 36. o589 (1942). 9. Bruson, U, 8. Patent 2,347,527, April 25, 1944. Original not seen. Abstracted in C. A., 39, 87 (1945).

10. Bruson, U. S. Patent 2,359,708, October 3- 1944. Original not seen. Abstracted in A., 3&72 (1945),

11» Bruson, U. S. Patent 2,372,808, April 3, 1945. Original not seen. Abstracted in A., 4623 (1945),

12. Bruson and Riener, J, Gh@B. Soc.. 64. 2880 (1942). 13, Bruaon and Hlener, J. Chem. 8oo.. 65. 18 (1943), 14, Bruson and Riener, J. M. Ch^. Joe., 65, 23 (1943),

15. Carotherg, U. 3. Patent 2,130,947, Septeasfeer 20, 1938. Original not seen. Abstracted in A., 9497 (1938), 96

16. Carothers, IJ. S. Patent I?, 130,523, September 20, 1938. Orlfclnal not aeen. Abstrr-cted in C. A., 9498 (1938).

17. Clifford and I^lohty, Canadian Patent 415,55=5, September 28, 1943. Orir^lnal not seen. Abstracted in C. A., 38, 979 (1944). 18. Colman smd Perkln, J. Chea. Soc.. 53, 191 (1888). 19. Condo, Hinkel, Fassero and Shrlner, J. M, Chem. Soc,. 230 (1937). 20. DpvldBon, J. Chem. Kducetion. 17, 81 (1940). 91, Delnert, J. prakt- Chem. ixjj, 52, 431 (1895). 22, Dreyfus, British Patent 313,651, Mf=rch 16, 1928, Original not seen. Abstracted In A,, 24. 1120 (1930).

23, Brushel and Dean, to, J, Sol. iS7» 486 (1913). 24, Druehel and Holden, M* J. Sol. /4/f 40^ 511 (1915), 26, Eijkraan, Cheta, Weekblad. 1, 421 (1904),

26, Eijkman, Chem. Weekbled. 2, 229 (1905),

£7, Sijkraan, Chen?, '"eekbled. 4, 191 (1907), Orif^inal not seen, Abatracfed in Qheffl". Z^ntr,, 1907 I, 1616, 28, Eijkraan, Chem. Weekblad. 4, 727 (1907). Orig-lnal not S6;en. Abstracted in Chem. Zentr.. 1907 II, 2045.

29, Fichter and Herndl, Helv. Chia. Acta, 14, 857 (1931). 30, Fichter and Schnider, Helv. Chitn. Acta. £5, 229 (1942), 31, Folkers and Adkins, J. Chem. Soc.. 54, 1145 (1932), 32, Freer and Perkin, J. Chero. Soc,. 51. 837 (1887), 33, Froebe and Hochstetter, Honatsh,. 23, 1088 (1902), 34, Giltaan, "Organic Chemistry, •• Second Edition, John Wiley and Sons, Inc., New York, 1943, p, 1838, 35, aoss and Ingold, J. Chem. Soc.. 127. 2779 (1925). 97

36, Hamonetj, Coapt. rencL.. 138. 260 (1901), 37, Hamonet, Bull, sop, chlm. /^. 33. 518 (1905),

38, Harrower, J. Qheni. 3oo«, 1931, 2052. 39. Hicklnbottom, "Reactions of Organic Compounds," Longmans, G-reen and Co., London, 1936, p. 198. 40. Hill and Adklns, J, Chem. Sag., 1035 (1938). 41, Hopff and Rapp, Crerman Patent 731,708, January 21, 1943. Original not seen. Abstracted in C. A., 38, 555 (1944).

42, Howells and Little, J. Chem. Soc.. 54. 2451 (1932).

431 Huber, J. M. Chem. Soc., 876 (1944).

ft 44, Hu<^el and Gelmroth, Ann.. 514. 233 (1934).

45, I. Q-. F'arbenlnd. A.-G-., French Patent 796,001, March 27, 1936. Original not seen. Abstracted In C. A., m, 5590 (1936). " " 46 Jonee and Powers, J. Chem. Soc., 2530 (1924). 47 Kervonen, Ann. Acad. Scl. Fennicae /A7. 10. No. 6 (1916). Orif^lnal not seen. Abstracted In Chem. Zentr.. 1919 III, 811. 48 Kllpi, 2' Physik. Chem., 86, 672 (1913). 49 Kimball, Jefferson and Pike, "Organic Syntheses," Coll. Vol. II, A. H. Blatt, Editor, John Wiley and Sons, Inc., New York, 1943, p. 284.

50 Kloetzel, J. Chem. Soc., 62^ 1708 (1940). 51 KoelBch, J. Am.. Chera. Soc.. 65. 437 (1943). 52 Kyrides and Graver, U. 3. Patent 2,368,366, January 30, 1945. Original not seen. Abstracted in C. A., 39, 4626 (1945). Lelniu, Ber., 70, 1051 (1937). Levene, Haller end Walti, J. Biol. Chera., 72, 593 (1927). Levene and Marker, J. Biol. Chem., 110. 329 (1935). 98

56. Llpp, Ber», 22, 2567 (1889),

57. Meyer and 8tamra, Ber.. 56. 1424 (1923). 58. Nazarov ff.nd Romanov, Bull, scad, sol. U. R. S. S., Clasee. sol, chlm.. 1940, 453. Original not seen. Abstracted In C. A., 3593 (1341). 59. Norrls and Arthur, J. to. Ghero. 3oc.. 62, 874 (1940).

60. Otterbach, "Organic Syntheses," Coll. Vol. I, Second Edition, H. G-llraan, Editor, John Wiley and Sons, Inc., New York, 1946, p. 290. 61. Pelomaa, Ann. Acad. 8ci. Fennlcae /A7. 3, No. 2 (1911). Original not seen. Abstracted In cKem. Zentr., 1912 II, 595. 62. Palomaa, Ann. Acad. Sol. Fennicae /I7. 4, No. 2 (1912). Original not seen* -Abstracted in Chero. Zentr., 1913 II, 1956.

63. Palomaa end Jaakkola, Ber.. 67. 949 (1934). 64. Paloiaaa and Kilpi, Ann. Acad. Sci. Fennlcae /a7, 2, No. 3 (1910), Original not seen. Abstracted in Chem. Zentr.. 1910 II, 1453. 65. Paloaaa and Tukklmaki, ^., §B, 887 (1935). 66. Pinner, Bei!'* 352 (1883).

67. Pinner, Ber., 3^, 1643 (1883).

68. Purdie and Marshall, J, Ghem. Soc.. 59, 474 (1891). 69. Radzissewski, 18, 355 (1885).

70. RehberR, Dixon and Fisher. J. Ara, Ghem. Soc.. 68, 544 (1946). "" 71. Ruzicka, 1367 (1917). 72. Sabetay, Bull, soc. chiia. /4?, 45. 534 (1929).

73. Schulz, U. S. Patent 2,287,537, June 23, 1943. Orifdnal not seen. Abstracted in A., 142 (1943).

74. Schwoegler and Adkins, J. Ghem. Soc.. 61, 34?9 (1939). 99

75. Qemnler, Per.. 39, 2852 (1906).

76. Shrjner and Puson, "The Systematic Identification of Orr;anlc Cost^ounda, " John Wiley and 3ons, Inc., New York, 1^935, p. 144. ?7. Sokolow, to., 150. 167 (1869). 70. Spiegel, Ber., 296 (1918).

79. Thomas, "Anhydrous Aliacinura Chloride in Orf:*anlc Chemistry," Reinhold Publishinp; Corporation, Ne\? York, 1941, pp. 67.3-678.

80. Thoiaa.s and Schuette, J. M. Ghem. Soc., 2324 (1931). 81. Tiiomas, Sctiuette and Cowley, J. M, Chern. 3oc.. 53. 3861 (1931).

82. Utermohlen, J. to. Chem. Soc.. CT, 1505 (1945). 83. ¥an Duzee aJid Adlvins, J. M. Gheia. Soc.. 147 (1935). 84. von Breun and Stuckenschnj^dt, Per., 56, 1724 (1923).

85. Walker, U. S. Patent 2,352,671, Ju3.y 4, 1944. Original not seen. Abstrr-cted in C. 223 (1945). 86. Yftiltinore, "Organic Chensistry, •• D. Van Noetrand Co-pany, Inc., New York, 1937, p. 447,

87. W'dtmore, liosher, Adgjis, Taylor, Olisx^in, V/eiaol and Yanko, i.' Chetn. Soc,. 66, 725 (1944). 88. Wied©3ann ^Jid Montroraery, J. Am. Che©. Soc., 67, 1994 (1945). ~ —

89. Vlmma and Adkins, J. Chem. Soc.. 4167 (1933). 90. Wislicenus, Ber., 3, 809 (1870).

91. ^^islicenua, to., 10 (1872).

92. ?/ohlr-emuth, Coapt. rend.. 158. 1577 (1914). 93. EelinBkii and Shuikin, Gpirmt. rend, aoad. sci. U. R. S. _3. /N. 1933, 60. .Original not seen. Ahstrrcted in C. A., 2002 (1934).